Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From the collection of the 



7 n 

m 

o Prefinger 



San Francisco, California 
2007 



LUME XXV NUMBER ONE 

JOURNAL 

of the 

SOCIETY OF MOTION 
PICTURE ENGINEERS 






JULY, 1935 



SHED MONTHLY BY THE SOCIETY OF MOTION PICTURE ENGfNEERS 



The Society of Motion Picture Engineers 
Its Aims and Accomplishments 

The Society was founded in 1916, its purpose as expressed in its 
constitution being the "advancement in the theory and practice of 
motion picture engineering and the allied arts and sciences, the stand- 
ardization of the mechanisms and practices employed therein, and the 
maintenance of a high professional standing among its members." 

The membership of the Society is composed of the technical experts 
in the various research laboratories and other engineering branches of 
the industry, executives in the manufacturing, producing, and exhibit- 
ing branches, studio and laboratory technicians, cinematographers, 
projectionists, and others interested in or connected with the motion 
picture field. 

The Society holds two conventions a year, spring and fall, at various 
places and generally lasting four days. At these meetings papers 
dealing with all phases of the industry theoretical, technical, and 
practical are presented and discussed and equipment and methods 
are often demonstrated. A wide range of subjects is covered, many 
of the authors being the highest authorities in their particular lines 
of endeavor. Reports of the technical committees are presented and 
published semi-annually. On occasion, special developments, such as 
the S. M. P. E. Standard Visual and Sound Test Reels, designed for 
the general improvement of the motion picture art, are placed at the 
disposal of the membership and the industry. 

Papers presented at conventions, together with contributed articles, 
translations and reprints, abstracts and abridgments, and other ma- 
terial of interest to the motion picture engineer are published monthly 
in the JOURNAL of the Society. The publications of the Society 
constitute the most complete existing technical library of the motion 
picture industry. 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV JULY, 1935 Number 1 



CONTENTS 

Page 

Progress in the Motion Picture Industry: Report of the 
Progress Committee 3 

Television and Motion Pictures A. N. GOLDSMITH 37 

The Theatrical Possibilities of Television H. R. LUBCKE 46 

Mechanographic Recording for Motion Picture Sound-Tracks . . 

J. A. MILLER 50 

The Kodachrome Process for Amateur Cinematography in 
Natural Colors L. D. MANNES AND L. GODOWSKY, JR. 65 

Introduction to the Photographic Possibilities of Polarized 

Light F. TUTTLE AND J. W. MCFARLANE 69 

A Device for Automatically Controlling the Balance between 
Recorded Sounds W. A. MUELLER 79 

Highlights of the Hollywood Convention 87 

Program of the Spring Convention at Hollywood 91 

Society Announcements 97 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREB, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St.. New York, N. Y. 



PROGRESS IN THE MOTION PICTURE INDUSTRY 
REPORT OF THE PROGRESS COMMITTEE* 

Summary. This report of the Progress Committee covers the year 1934. The 
advances in the cinematographic art are classified as follows: 

(I) Cinematography; (II) Sound Recording; (III) Sound and Picture Repro- 
duction; (IV) Film Laboratory Practice; (V) Application of Motion Pictures; 
(VI) Publications and New Books; Appendix A General Field of Progress of the 
Motion Picture Industry in Great Britain; Appendix B General Field of Progress 
of the Motion Picture Industry in Japan. 

INTRODUCTION 

When the Committee undertook its annual task of preparing the 
Progress Report for 1934, it looked at first as though there might 
be a considerable dearth of new material. However, when the 
Committee settled down to active work, the prospects for a good 
report seemed brighter than at any time during the past three years, 
and when the contributions from the individual members of the 
Committee were assembled into the following general report, many 
new items of interest and of progress were seen to have come into 
evidence during the past year. 

It is true that there were no very outstanding achievements in the 
cinematographic art during 1934, but there was much consistent 
progress in improvements of equipment and technics of operation. 
It is of interest to note, for example, that transparency background 
work nearly attained perfection during the past year. In the field 
of studio illumination the use of a new mercury arc is reported to 
have great possibilities. Considerable progress is reported in the 
use of camera accessories to facilitate the photography of difficult 
shots, thereby adding more realism to the scene. 

The use of color in motion pictures continued to receive consider- 
able emphasis during the past year, some very popular cartoons and 
two-reel variety subjects having been made exclusively in color. In 
the field of sub-standard films considerable progress has been noted 
in recording and reproducing sound, but ultimate progress in this 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 



4 REPORT OF PROGRESS COMMITTEE [J. S. M. P. E. 

field is hampered by the inherent difficulties of securing a sufficiently 
wide frequency range. A great many new substandard projectors 
were announced during the year. 

In the field of sound recording, progress in the United States was 
almost at a standstill during the year while the battle over the 
validity of the Tri-Ergon patents was raging in the courts. In 
spite of this, however, considerable progress is reported in the gradual 
extension of wide-range and high-fidelity recording in the theaters 
of the country. Audiences appear to be becoming more sound- 
conscious, and the reaction of the public to the excellent recording 
in Columbia's One Night of Love augurs well for high-grade produc- 
tions of this type in the future. 

In the field of sound reproduction, the successful introduction of 
all a-c. operated theater equipment during the past year seems to 
offer the possibility of improved and simplified equipment at a con- 
siderable saving to the exhibitor. Of interest, also, to the exhibitor 
is the introduction of copper oxide three-phase full-wave rectifiers 
for supplying power for projector arcs in theater booths. In the 
field of film laboratory practice there is little progress to report during 
the past year, and the same is true in the field of application of motion 
pictures. 

The committee is happy to be able to include a very excellent report 
on progress in the motion picture industry in Great Britain, as out- 
lined in Appendix A. The very substantial progress of the motion 
picture industry in Great Britain is entirely consistent with the re- 
markable success of British-made pictures in the United States and 
other countries during the past year. A separate report listed in 
Appendix B covers progress of the motion picture industry in Japan. 

The committee wishes to thank the following firms for supplying 
photographs for use in the report: Paramount Productions, Inc., 
W. C. Hollins Electric & Engineering Co., Eastman Kodak Co., 
Bell & Howell Co., RCA Manufacturing Co., Bell Telephone Labora- 
tories, Inc., and Electrical Research Products, Inc. 

J. G. FRAYNE, Chairman 

M. ABRIBAT R. E. FARNHAM R. F. MITCHELL 

L. N. BUSCH E. R. GEIB G. F. RACKETT 

A. A. COOK G. E. MATTHEWS P. R. VON SCHROTT 

R. M. CORBIN H. MEYER S. S. A. WATKINS 

J. A. DUBRAY V. E. MILLER I. D. WRATTEN 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 5 

SUBJECT CLASSIFICATION 

(I) CINEMATOGRAPHY 

(A ) Professional 

(1) General 

(2) Films and Emulsions 

(3) Cameras and Accessories 

(4) Studio Illumination 

(5) Color 

(B) Substandard 

(1) General 

(2) Films and Emulsions 

(3) Cameras 

(4) Projectors 

(5) Color 

(6) Sound Recording 

(7) Sound-Film Projection 

(II) SOUND RECORDING 

(1) General 

(2) Recording Equipment 

(3) Accessories 

(III) SOUND AND PICTURE REPRODUCTION 

(1) Sound Equipment 

(2) Projectors and Accessories 

(IV) FILM LABORATORY PRACTICE 

(V) APPLICATIONS OF MOTION PICTURES 

(1) Education 

(2) Race Timing Devices 

(VI) PUBLICATIONS AND NEW BOOKS 
APPENDIX A 

General Field of Progress of the Motion Picture Industry in Great 

Britain 
APPENDIX B 

General Field of Progress of the Motion Picture Industry in Japan 

(I) CINEMATOGRAPHY 

(A ) Professional 

(1) General. Strictly speaking, there were no outstanding 
developments in cinematography in 1934, but a general forward 
movement in several phases of photographic work indicates a wide- 
spread interest that presages marked advancement in the near future. 



6 REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 

Perhaps the most needed piece of equipment in the industry is 
a silent camera. The leading camera manufacturers are working 
hard, but are not yet able to supply the industry with a camera that 
fulfills the sound and weight requirements of the studios. The 
Silent Camera Committee of the Academy of Motion Picture Arts 
and Sciences has been gathering material and data that should prove 
helpful when properly catalogued and made available to those 
interested. 

In the meantime each studio has been working industriously 
toward further perfecting their "blimps" by making them quieter 
and of lighter weight. Some of the "blimps" are sufficiently quiet 
to meet the most stringent demands of the sound departments, but 
the weight has been reduced comparatively little, leaving a real need 
for a one-man unit. 

The operation of the "blimps" has been simplified greatly by adding 
follow-focus and parallax-correcting finders, electrical synchronizers, 
more accessible matte and diffuser accessories, and better free-heads. 
Such improved blimps may be operated as quickly as the old silent 
cameras, with the one exception of changing set-ups, requiring addi- 
tional man-power. 

Projection printers have grown from two old Bell & Howell cameras 
on a discarded lathe bed into a complicated mechanism that will do 
unbelievable things to a negative after the cameraman has turned it 
into the laboratory it has become the Aladdin's Lamp of the 
cameraman's wildest dreams. Zoom lenses have made great ad- 
vances during the past year and, although not yet perfected, a zoom 
lens can be expected in the near future that will operate at an //2.3 
speed from a 35-mm. angle to a 150-mm. angle, making possible a 
single lens doing the work of at least six lenses of the present type. 

Projection or transparency background work has nearly attained 
perfection during the year. The hot spot, though it still exists, has 
become a minor difficulty; perfect synchronization and matched 
lighting have blended composites into a much more beautiful whole 
than was ever before possible. Improved technic has widened the 
scope of transparency projection until "location trips" have become 
one-man jobs, the cast restricting their trips to the studio stage, a 
large piece of ground glass furnishing the requisite locale of desert, 
mountains, or foreign countries. Excellent examples of this work 
are shown on Figs. 1 and 2. 

Lighting equipment is being improved in the incandescent field 



July, 1935] PROGRESS IN MOTION PICTUR.E INDUSTRY 7 

by designing the reflectors so that, instead of following parabolic 
or mathematically known curves, they describe unrelated curva- 
tures. By this means the shadow of the filament or "ghost," is 
eliminated and the outer field of illumination is blended until it loses 
its defining circle of light, thus simplifying group lighting. 

There seems to be a renewed activity in stereoscopic research, but 
nothing really commendable has come to the attention of the Com- 
mittee. 

Considerable agitation has developed relative to a change of film 
speed from 24 to 20 frames per second, the object being to effect a 




FIG. 1. 



Projection background shot (from The Menace, a Paramount pro- 
duction) . 



large saving in the cost of film. Its proponents claim that neither 
sound nor picture will suffer by such a change, but that production 
costs will be lessened by at least half a million dollars a year. The 
burden of argument against such a change lies largely with the sound 
departments. Photographically, nearly every picture released has 
speed changes varying from extremely low speeds to speeds several 
times the normal, without infringing upon the laws of persistence of 
vision. 

Colored glass of known filter values has been substituted for clear 
glass in several large combination exterior-interior sets with excellent 
results, enabling the cinematographer to "shoot" from an interior to 



8 REPORT OF PROGRESS COMMITTEE IJ. S. M. P. E. 

an exterior without "burning up" the exterior scene. This is not new 
in principle, but cheap glass may now be obtained, so that its use 
becomes practical. 

(2) Films and Emulsions. The high standard of emulsion sensi- 
tivity, freedom from halation, and the wide range of color sensitive- 
ness established several years ago for panchromatic films for general 
use have been maintained during the past year. New materials 
have been limited chiefly to those designed for special processes. A 
fine-grain panchromatic film known as Background Superpan was 
added to the list of materials used for projection background pho- 




FIG. 2. Projection background shot (from Mississippi, a Paramount pro- 
duction). 

tography. Other new emulsions included a high-contrast film for 
making travelling mattes for special process photography, a medium- 
contrast fine-grain duplicating negative film, and two fine-grain high- 
contrast sheet films for commercial photography. Improved emul- 
sions sensitized to the infrared were also announced. The charac- 
teristics of Agfa Superpan negative motion picture film were dis- 
cussed in a paper by Arnold. 1 

A rather novel emulsion for making direct duplicates by direct 
printing was described by Koch. 2 Exposures varying from 4 seconds 
to 6 minutes, using a 40-watt bulb at 25 cm. were found necessary 
because of the low sensitivity of the emulsion. Schweitzer 3 reported 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 9 

that this film was slightly fogged in manufacture, the latent image 
being destroyed upon exposure to light. Many articles were pub- 
lished during the year in the German technical journals dealing with 
fine-grain development. One of the many articles dealing with this 
problem is that by A. von Barsy, 4 who states that in practice the 
fine-grain development has proved to be of value, whereas Luppo- 
Cramer 5 expresses the opinion that, if the developer is sufficiently 
diluted and the development time correspondingly increased, any 
formula can be used for the fine-grain development. 

Of especially great interest are those articles 6 that deal with new 
methods for determining the graininess of photographicfi 1m. I. Eg- 
gert and W. Kiister of the I. G. Farbenindustrie propose to deter- 
mine the graininess by means of the Callier effect, and their proposal 
caused A. Narath of the Klangfilm Laboratory to investigate 
this method. 7> 8 

A comprehensive group of papers 9 ' 10 by several Russian scientists 
dealing with various aspects of emulsion manufacture were published 
by the Kino-Photo Institute of Moscow and the Photo-Kino Indus- 
try. The subjects discussed included optical sensitizing, emulsions 
without ammonia, aging of emulsions, ripening, effect of washing 
upon light sensitivity, etc. 

An investigation 11 was initiated by Sheppard, Wightman, and 
Quirk upon the temperature coefficient of photographic sensitivity. 
Among other results, it was found that the optical sensitizing effect 
produced by dyes persisted at 190C. but was affected in much 
the same degree as the blue-violet sensitivity of the less sensitive 
grains. Webb 12 reported on the effect of temperature upon the 
reciprocity law failure in photographic exposure and stated that at 
very low intensities, such as are occasionally used in astronomical 
research, speeds of certain emulsions can be increased several-fold by 
lowering the temperature. 

The photographic activity of gelatin was shown by Rabinovitch 
and Titoff 13 to be proportional to its content of labile sulfur as 
determined by the method of Sheppard and Hudson. Bekunoff 14 
reported that hydrolysis of gelatin had very little effect upon the 
photographic qualities of emulsions made from it. 

The output of raw motion picture films in Russia was increased 
considerably in 1934 over 1933, as part of the expansion program 
of the second five-year plan. 15 During the year, the first plant for 
manufacturing film was established in Denmark. 16 



10 REPORT OF PROGRESS COMMITTEE [j. s. M. p. E. 

Patent protection was granted 17 on several additional schemes for 
introducing anti-halation layers in films. Pederson 18 patented a 
method of reenforcing the perforation area of motion picture film 
which consists in attaching armored strips having sprocket holes 
spaced by the same distance as the film perforations. 

(3) Cameras and Accessories. The year 1934 saw little in the 
way of new professional camera equipment. Any advances could 
be considered only as modifications of previous designs, rather than 
as new departures in design or technic of operation. From England, 
W. Vinten, Ltd., announced the adaptation of a 40-inch Dallmeyer 
lens to a Vinten Model H camera, 1000-ft. magazines and motor. 
The lens is mounted upon a special platform which also carries the 
camera, and the whole is carried upon a gyroscopic tripod. This 
camera has been extensively used in newsreel work. 

A new departure in blimp construction is described in American 
Cinematographer. lg It consists essentially of a silenced Mitchell 
Camera rebuilt as an integral part of a light sound-proof housing. 
All controls and meters are operated or are visible from outside the 
case, which needs to be opened only for loading. A single lens, rather 
than a turret, is utilized. The weight of the outfit is stated to be 
116 pounds when loaded and ready for use. Bell & Howell have 
announced 20 that their portable 35-mm. Eyemo Camera has been 
adapted so as to be used interchangeably for regular motion picture 
work or for taking single exposures at set intervals for aerial map- 
ping work. 

Paramount Productions, Inc., have designed the tripod perambu- 
lator illustrated in Fig. 3. It is elevated electrically to a height of 
eight feet from the floor, controlled by a rheostat operated by the 
cameraman. It eliminates tripod, high-hats, perambulators, and 
all other blimp supports except for high parallel shots. It is claimed 
that the use of this device speeds up production at least an hour per 
day. The blimp shown in the illustration operates entirely from 
outside controls, the follow-finder and follow-focus being controlled 
from a calibrated lens. 

The year 1934 has had nothing startling to offer in the way of 
new camera lenses. The number of patents issued 21 indicates 
that developmental work is going along at its usual pace in this 
field, but that few of these new constructions have reached the 
market in the form of finished objectives. R. Kingslake has written 
a history of the development of the photographic lens 22 that should 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 11 

be of value to all those interested in lens design. J. A. Dubray has 
published an article on chemical focus in cinematography, 23 and 
McFarlane has described the use of supplementary lenses with 
16-mm. cameras. 24 The automatic control of diaphragm openings 
in photography by means of a photo-cell in the camera finder is pro- 
posed in a recent patent. 25 




(Courtesy of Paramount Productions, Inc.) 
FIG. 3. Camera dolly crane. 

(4) Studio Illumination. One of the lamp manufacturers (General 
Electric) has developed a new type of tubular lamp known as the 
Lumiline, which consists of a tubular bulb 1 y 4 inches in diameter and 
either 12 or 18 inches long, with metal caps sealed to the glass at the 
ends of the bulb, through which contact is made. The socket con- 
sists of a small cap containing the contact, covered with insulating 



12 



REPORT OF PROGRESS COMMITTEE 



[J. S. M. p. E. 



material which is "buttoned" onto the ends of the lamp. When 
lamps are placed in line, the break in the light-source is scarcely more 
than 1 / 2 inch, thus making practically a continuous line of light. 
The lamp is intended for general illumination as well as for decorative 
lighting, and is available in a number of colors in 40- and 60-watt 
ratings. 

In last year's report it was mentioned that experimental work was 





(Courtesy of Mole-Richardson, Inc.) 
FIG. 4. 1000-ampere choke-coil. 



(Courtesy of Eastman Kodak Co.) 

FIG. 5. Kodascope model L, 



being done upon a high-intensity mercury vapor lamp. This lamp 
is now commercially available in the 400-watt size with 14,000 lumens' 
output. It will undoubtedly prove to be of value in the motion 
picture industry, particularly for process work. Its technical fea- 
tures were described by Buttolph. 26 

Two other lamps have been developed which may be of interest 
to the motion picture industry for certain lighting effects. The first 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 13 

is a three-light lamp, containing two filaments. It is provided with 
a special three-contact mogul-screw base, making it possible, by 
proper switching, to obtain three levels of illumination. The lamp 
is available in the 150- to 200-watt and 200- to 300-watt sizes. The 
other lamp is a high-speed flashing lamp of 1000- watt rating, the 
bulb of which contains hydrogen instead of the usual argon or 
nitrogen. It is possible to transmit flashing signals at twice the speed 
that was possible with the mechanical shutter. 

The Corning Glass Works has developed a special blue-glass filter, 
known as their Lunar White No. 570 which, when used with the 
Movieflood lamp announced a year ago, gives almost perfectly white 
light. This combination of lamp and filter is particularly applicable 
for color motion picture photography, which, in general, requires 
substantially equal quantities of the three primary colors. The 
details of the glass and its application to studio lamps was discussed 
in a paper by R. E. Farnham. 27 

The interest in carbon arcs for studio lighting is quite pronounced 
at the present time, due to the new Technicolor pictures. In order 
to render the operation of the arcs sufficiently quiet for sound pic- 
tures, L. Kolb, of M-G-M, recently built a number of choke-coils of 
1000-ampere capacity. Fig. 4 shows similar coils built by Mole- 
Richardson Co., Inc. Each coil contains 300 feet of 1,000,000 
circular mil copper cable, making 36 turns, and has an air core. 
Such a choke-coil has the advantage of taking care of a large number 
of arc lamps from its position near the power-house. The ordinary 
choke-coil, made for individual lamps, must be carried to the motion 
picture set and located near the lamp it serves. 

Along this same line is the announcement that the W. C. Hollins 
Electric and Engineering Company of Los Angeles has developed a 
dry type of electrolytic condenser to replace the regular electrolytic 
condensers in present use in the West Coast studios. Each unit 
weighs approximately 25 pounds, and has a capacity of 2500 
microfarads. One unit is sufficient for each end of a generator. 
These condensers are used in conjunction with choke-coils to 
eliminate commutator ripple. The advantages claimed for the "dry 
type" condenser are lightness of weight, no care is needed, lowness 
of price. They have recently been installed in several of the 
West Coast motion picture studios. 

(5) Color. A renewed interest in color processes was apparent 
among several of the producing organizations. A number of two- 



14 REPORT OF PROGRESS COMMITTEE [j. S. M. P. E. 

color cartoons were made by various companies in the United States 
and Europe. The Caspar process was used in making several three- 
color cartoons shown at Berlin and London last fall. This process 
utilizes a film composed of three dyed emulsion layers coated upon 
one support. The film is exposed successively to each one of a set 
of color separation positives. The three silver images are developed, 
fixed and washed, and the film is then passed through a bath con- 
taining thiocarbamide, by means of which the dye in each layer is 
rendered colorless in conjunction with the silver in the layers, the 
amount of bleaching being proportionate to the amount of the silver 
deposit. The silver is then rehalogenized and fixed out. There 
remains a subtractive three-color picture image composed of dyes. 28 

The improved Technicolor process was used in making a large 
portion of the products supplied by Walt Disney. 29 A two-reel 
color picture entitled La Cucuracha was released in September in the 
United States, made by the three-color Technicolor process, and a 
complete feature was announced for release during 1935. These 
represent probably the first three-color subtractive pictures, apart 
from cartoons, made for commercial showing in this country. 30 

Having found that silver chloride treated with a solution of a 
diazotate can be readily converted into silver diazotate, the I. G. 
Farbenindustrie Aktiengesellschaft worked out a method of making 
color prints from silver images. 31 

(5) Substandard 

(1) General. The past year was marked by decided progress and 
interesting developments in this field, principally due to progress 
made in manufacturing and supplying equipment for recording and 
reproducing sound on substandard film; the use of substandard film 
sizes for motion pictures has been largely extended from the former 
amateur field to that of the commercial 32 and educational field. 

Further improvements made in manufacturing fine-grain emulsions 
and the general introduction of special fine-grain developing formulas 
have resulted in an increased interest in the possibilities of the nega- 
tive-positive processing method in competition with the reversible 
process. 

While the 16-mm. size is rather exclusively preferred by American 
users and film manufacturers, it will be found that in Europe the 
market for 9.5-mm. film, originally suggested in France, is given 
much attention, as well as that for 17.5-mm. film. The newest 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 15 

development stimulated by the introduction of the 8-mm. projector 
by Eastman in the United States has led to the manufacturing of 
8-mm. reversible film in Germany. 

(2) Films and Emulsions. Several new 16-mm. films have been 
produced in the United States. Plenachrome, a new 16-mm. ortho- 
chromatic reversible film, is manufactured by the Agfa Ansco Corp. 
The same firm introduced a fine-grain 16-mm. panchromatic nega- 
tive film. The Eastman Kodak Co. supplied a low-priced pan- 
chromatic 16-mm. reversible film. 

In Germany the I. G. Farbenindustrie (Agfa) has issued a new type 
reversal film called Isopan Umkehrfilm, the speed of which is greater 
than that of the Novapan film, a product of the same firm, and which 
also has a finer grain. 33 The Agfa "Ozaphan" film was brought upon 
the market during the last year. Agfa have already built up a great 
library of these films, and every month the amateur can buy news- 
reels taken on this material at a very low price. A rather complete 
report covering the photographic characteristics of European sub- 
standard films recently developed, including filter factors, has been 
published in England, naming the films manufactured and distributed 
by Eastman, Agfa, Ilford, Gevaert, Pathe", and Bolex. 34 

(3) Cameras. A great number of new amateur cameras were in- 
troduced or improved during the past year. A 16-mm. camera 
driven by an electric motor has been marketed by Amigo, Berlin. 35 
A camera for 16-mm. cassettes has been introduced by Niezoldi and 
Kramer. 36 The same firm announces the first 8-mm. camera pro- 
duced in Germany. 37 Another 8-mm. model has recently been 
manufactured by the French firm, Establissements Emel. Keystone 
also constructed an 8-mm. camera. 38 

A new Zeiss 16-mm. camera called the Ikon Movikon camera has 
been developed, using daylight spools. It can be operated at various 
speeds and is probably the first cinematographic camera which has a 
distance meter, coupled with the taking lens. The sector of the 
rotating shutter is adjustable. The Siemens camera, which is already 
well known, shows some further improvements. Model C has four 
speeds and is now fitted with a Meyer-Siemens anastigmat//1.5. 

(4) Projectors. A large number of new or improved models have 
been offered by manufacturers in the United States and abroad 
furnishing greater illumination, mechanical flexibility, and compact- 
ness. Eastman announced the Kodascope L (Fig. 5) for 16-mm., 
which is supplied with various lenses and lamps up to 750 watts. 39 



16 



REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 



The Bell & Howell Co. met the trend for greater wattage by 
manufacturing Model 130 with a 1000- watt lamp. Improvements 
incorporated in this model consist of a special cooling system em- 
ploying two fans and an air-conditioned panel through which the 
film is driven and is humidified. The optic consists of a Cook lens 
//1.65 40 (Fig. 6). A similar model, 129, with a 750-watt lamp, is 




FIG. 6. Bell & Howell 16-mm., 1000-w. projector. 

also available. 41 In both models new "stream-line" styles have been 
developed. 

The Victor Animatograph Corp. announced the 16-mm. Super 
Hi-Power projector with a 750-watt lamp. 42 Abroad, Lytax, in 
Germany, completed the Super P projector, a rather universal 
equipment, as it allows projection of 8-, 9.5-, 16-, and 1.75-mm. 
films and, in addition, projection for 5 X 5 cm. stills. 43 An improved 
modelof the well-known Niezoldi and Kramer line has been announced 
which is called the Niezo II. S. 44 Agfa have issued a new projector, 
the Movector Super 16, which can be used for projecting 16-mm. 
reversal film, lenticulated film and Ozaphan film. 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 17 

Of special interest are the arrangements intended to reduce the 
projector noise. This is achieved by directly coupling the ventilator 
and the rotating shutter with the axis of the motor. The rotating 
shutter turns twice for each frame; for sound projection it is used as 
a one-blade shutter, and for the projection of silent pictures it works 
with two blades at a speed of 16 frames a second. 

The Bell & Howell Company announced a series of reels of varying 
capacities: 1600, 1200, 800 feet for 16-mm., and 200 feet for 8-mm. 
films (Fig. 7) . All reels are equipped with prongs which automatically 
hold the film at the central core. The interesting innovation consists 
in using spring steel for the reels instead of the conventional aluminum 




(Courtesy of Bell & Howell Co.) 
FIG. 7. Spring steel reels. 

insuring against warping or distortion even if the reels are submitted 
to accidental shock, at the same time maintaining lightness and 
rigidity. 

(5) Color. Subtractive two-color processes have been used to 
some extent commercially but no decided progress or new principle 
has been made known in this field. In England the Spicer-Dufay 
color process has been announced as available also for 16-mm. 
reversible film. Several technical papers describing this process, 
which employs a color mosaic screen, and includes processing for- 
mulas as well as methods applicable for making duplicates, have 
been published. 45 A demonstration of the Dufaycolor regular mosaic 



18 REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 

process on 16-mm. film was given in April, 1934, at the Spring 
Meeting of the Society. The screen rulings were about 1000 to 
the inch. 46 

During the year, Agfa introduced an improved color-film of in- 
creased sensitiveness and balanced for daylight exposure without a 
filter. 47 This was made available in amateur camera sizes and as 
35-mm. motion picture films for Leica, Contax, and similar cameras. 

(6) Sound Recording. Although in the United States the variable- 
width recording system is given preference in connection with sub- 
standard film, in Germany the variable-density system is found to 
be in greater use. 48 Since the singly perforated 16-mm. film became 
standard, the leading film manufacturers have supplied the market 
with the corresponding types. Direct recording on 16-mm. film 
has been the subject of much study and research, but in spite of all 
efforts the problem has not yet been satisfactorily solved. The 
RCA Manufacturing Company has done outstanding pioneering 
work in this field. Other than the RCA 16-mm. sound camera, 49 
and the Berndt-Maurer 16-mm. sound recorder, 50 no new apparatus 
has been announced in the United States. 

The problem of reducing 35-mm. sound records to 16-mm. for 
reproduction has been given increased attention during the year. 
The results and principles of the investigations in this field have been 
published in a number of papers and indicate that the problem has 
come very close to a practical solution. 51 

The various possibilities of either reaching the final composite 
16-mm. sound-and-picture print by optical reduction or by means 
of re-recording from the 35-mm. original on to 16-mm. film have 
been thoroughly studied. Both methods have their advantages and 
disadvantages which, balanced against each other, have most of the 
authors and investigators deciding in favor of optical reduction. In 
connection with this work the question of suitable printing equipment 
is naturally being investigated also, resulting in the announcement of 
improved printers for optical reduction. Thus, A. F. Victor 52 and 
G. A. Busch 63 described a continuous optical reduction printer; in 
England, W. Vinten, Ltd., 64 also announced an optical reduction 
printer from 35-mm. to 16-mm. film; and in Germany, Arnold and 
Richter. 65 A 16-mm. contact printer has been manufactured by 
Berndt-Maurer, 56 and by Phillips Laboratory in New York. 87 

(7) Sound- Film Production. The progress achieved in perfecting 
methods for successfully producing 16-mm. sound-film motion 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 19 

pictures, and their importance in the field of commerce, industry, 
education and entertainment, has encouraged an additional number 
of manufacturers to market new sound projectors. The following 
sound-film projectors have been announced : 

Ampro (16-mm.); 68 Electronic Devices Corporation (Edco) 
(16-mm.); 59 De Brie (16-mm. and 17.5-mm.); 60 British Thompson- 
Houston Corporation (16-mm.); 61 Agfa, Lichton Movector (16- 
mm.); 62 Siemens Lichton (16 mm.); 63 Donelli (Italy) (17.5 mm. with 
single center perforation); 64 Sales Producers, Ltd., London 
(16-mm.); 65 Photokinox, Zeiss Ikon (16-mm.). 

(II) SOUND RECORDING 

(1) General. There was no outstanding accomplishment in the 
field of new sound recording equipment during 1934. This is prob- 
ably due to the fact that major equipment suppliers were embroiled 
in litigation over the Tri-Ergon patents and hesitated to go ahead 
with new equipment lines until these patents were finally settled 
one way or the other. Both the Tri-Ergon patents having been 
held invalid by the Supreme Court, the decks should be cleared for 
action this year and it is hoped that some worth-while things may 
be announced a year from now. 

Considerable attention was given by the public to the sound 
recording in Columbia's One Night of Love, featuring Grace Moore. 
This picture marked the initial attempt to use the new vertical cut 
recording system in sound pictures. The songs and orchestral selec- 
tions were first recorded on wax and later transferred to film in the 
re-recording process. This gave a final film indistinguishable from 
an original film recording and superior to a film-to-film re-recording. 
The successful use of vertical cut by Columbia has stimulated its use 
in other West Coast studios. 

(2) Recording Equipment. RCA Photophone announces a new 
accurate monitoring device for use by the mixer. It consists essen- 
tially of a strip of neon lamps. The device has been tested over a 
period of several months in an eastern studio, and has been found to 
be very satisfactory. It indicates instantaneous levels with an 
accuracy permitting full modulation of the track without overshoot- 
ing. It is expected that the device will prove to be effective in pre- 
venting over-modulation when recording musical selections in which 
it is desired to utilize the full range of possible modulation without 



20 



REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 



overshooting. Modern reproducing equipment requires that this 
be prevented. 

A new system of recording on film was demonstrated which should 
serve to improve the fidelity and reduce the background noise of the 
extended frequency range of musical recordings. The new method 
can be employed for original recordings that are to be re-recorded 
for making final negatives. The original recording may be made 
either in the form of a positive which is almost completely exposed 
throughout the sound-track area, or in the form of a negative from 
which prints can be made for reproduction. The sound-track is 
divided into two parallel sections, each being exposed on only one- 
half of the sound-wave. One portion of the sound-track is then a 
record of negative half -cycles, and the other of the positive half- 
cycles. The form of the track is illustrated in Fig. 8. For silent 




() (0) 



FIG. 8. RCA double sound-track. 



intervals, there is practically no exposure when making positives. 
No biasing system is required, as with the noiseless recording systems 
used at the present time; it will still be required for making nega- 
tives from which release prints are to be made. 

In reproduction, the light on each track is impressed upon separate 
photo-cells or separate cathodes of a single cell. The cathodes are 
connected to opposite ends of a transformer winding. Imperfections 
in the tracks affecting both in the same way are balanced out. This 
eliminates distortion which might otherwise be present if reproducing 
from an original recording. 

The volume range with this type of recording is 55 db. without 
audible background noise, when used with systems having an effective 
frequency range to 9500 cycles. Only minor modifications of present 
recorders and reproducers for re-recording are required to utilize 
this sytem. 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 21 

There has been an increasing demand for a small microphone that 
can be more easily concealed on the set in scenes that require that 
it be within the camera angle. This demand is being met with the 
new non-directional moving-coil microphone shown in Fig. 9, which 
was recently designed by the Bell Telephone Laboratories. The 
microphone is not only small and convenient to use, but has an 
exceptionally flat response characteristic over the audible frequency 
range that is only slightly affected by direction. 

A sound-track engraving apparatus for engraving variable-width 
sound-tracks upon film was reported in Electronics. 

Warner Brothers-First National report marked success in using 
a resonant shunt across the light- valve. It is claimed that the shunt 





(Courtesy of Bell Telephone Laboratories, Inc.) 

FIG. 9. Non-directional dynamic microphone. 

practically eliminates transients and results in a recording that is 
much cleaner and free from overloads. 

(3) Accessories. An apparatus for measuring and analyzing 
flutter in recording and reproducing machines has been developed 
by the Electrical Research Products, Inc. It will measure frequency 
variations over a range from about 0.04 to 3 per cent, the indicated 
percentage being practically independent of the flutter rate. It is 
compact and readily portable, being contained in two small cases 
weighing about thirty-five pounds each. It is operated from a 
110- volt, 50- or 60-cycle power source, and is adaptable for use in 
the field as well as in the laboratory. 

An improved design of the Electrical Research Products, Inc., 
noise meter has recently been announced. The equipment is about 
the same as before, except that provision has been made for measure- 



REPORT OF PROGRESS COMMITTEE 



[J. S. M. p. E. 



ments where a flat frequency response is required instead of the 40-db. 
loudness ear response. In order to accomplish this, it has been 
necessary to redesign the amplifier to have a uniform response, and 
then provide the ear-weighting characteristic in the form of a sepa- 
rate network that may be switched in or out of the circuit by oper- 
ating a key. In order to obtain uniformity throughout the system, 
the standard Western Electric moving-coil microphone has been 
substituted for the special noise transmitter. 




(Courtesy of Electrical Research Products, Inc.) 

FIG. 10. All a-c. theater sound projector. 
(UI) SOUND AND PICTURE REPRODUCTION 

(1) Sound Equipment. While no radical innovations in sound 
reproducing equipment have been introduced during the past year, 
consistent progress in simplifying and cheapening equipment has 
been made. 

The Electrical Research Products, Inc., developed during 1934 a 
new sound system for theaters under 600 seats, to be known as the 
No. 5 type (Fig. 10). The outstanding features of the equipment 
are increased power, improved frequency response, operating sim- 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 



23 



plicity, and reduced installation expense. The important equipment 
components are the well-known 206 type reproducer set, a compact 
amplifier-rectifier assembly for front- wall mounting between pro- 
jectors, and a composite stage loud speaker consisting of two units in 
a horn-baffle arrangement. The loud speaker system is furnished in 
two different types, the choice depending upon the shape of the 
auditorium. 

The 86-A amplifier has recently been developed by the Western 




(Courtesy of Electrical Research Products, Inc.) 
FIG. 11. All a-c. theater sound amplifier. 

Electric Company for theater reproduction. The principal features 
of this amplifier which distinguish it as an advance in the art are its 
high gain (97 db.) with practically a flat characteristic from 40 cycles 
to 10,000 cycles, its high power output capacity (15 watts), and its 
small size. The amplifier proper is of the chassis type, weighing 
only 35 pounds. The amplifier is entirely a-c. operated, and has 
power supply for auxiliary amplifiers. It is normally mounted with 
auxiliary equipment in an attractive aluminum, gray-finished cabinet 
19 X 19 X 10 5 /s inches in size, arranged for rack or wall mounting 
(Fig. 11). 
RCA Photophone announces that theaters have continued to 



24 



REPORT OF PROGRESS COMMITTEE [J. S. M. P. E. 



install their high-fidelity equipment. There are over 1000 installa- 
tions in the United States. The later musical recordings having a 
greater difference in the recorded level of speech and music, and 
requiring a greater gain in power output for adequate reproduction, 
are impressing upon exhibitors the need for improved reproducing 
equipment that is free of system noise at the higher gains required 
and capable of reproducing loud passages without distortion. 





FIG. 12. RCA loud speaker system. 

The same company announces also that a new stage speaker has 
been made available for theater use (Fig. 12). It is especially suitable 
for theaters giving stage performances, because the entire assembly 
can be flown with the screen. In all installations it is especially 
convenient; in many installations, the assembly can be placed 
directly upon the floor behind the screen. When it is necessary to 
elevate the speakers for proper sound distribution, the assembly 
can conveniently be placed upon a simple platform. The speaker 
assembly consists of a large folded horn for reproducing frequencies 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 25 

below 125 cycles, and two or three small speakers or directional 
baffles for the range from 125 to 9500 cycles. The large folded horn 
serves to support the smaller units, which are hinged to the large 
horn in such a way that they can be conveniently tilted to the 
required angle. 

(2) Projectors and Accessories. H. A. DeVry, Inc., announce a 
projector "made from the ground up" for both sound and picture. 

The DeVry engineers have designed both picture and sound 
mechanisms as a unit, and have thus been able to eliminate many 
unnecessary parts. The silent-chain drive is substituted for meshed 
gears; rear barrel shutter is incorporated in the stock equipment, 
and the Robertson fly-wheel is utilized for filtering out vibrations. 

The unit is furnished in several models, for Mazda lamp, or for 
either high- or low-intensity arcs. All machinery is inclosed in a 
dust-proof metal case, the projector presenting a handsome stream- 
line effect from floor to magazine. Controls are accessible, however, 
from the outside. The unit may be used in the largest as well as in 
the smallest theaters. 

The General Electric Company has produced a three-phase full- 
wave copper oxide rectifier for supplying power to projector arcs in 
theater booths. Two sizes are available. The smaller is designed 
for use with the 6- and 7-mm. trim and will deliver 40 to 50 amperes 
at an arc voltage of 30 to 35 volts. The larger unit is designed for the 
6.5- and 8-mm. trim and will deliver 50 to 65 amperes. These ratings 
are in accordance with the standards established by the Projection 
Practice Committee of the S. M. P. E. 

The over-all efficiency of either of these units is better than 70 
per cent. According to findings of the Projection Practice Com- 
mittee, the high efficiency of the polyphase type of rectifier will show 
a saving of 5 to 10 cents an hour in the cost of current as compared 
with other types of power supply. This means that the rectifier 
will pay for itself in one or two years in current saving alone. 

The Morelite Company, Inc., of New York have placed upon the 
market a reflector arc lamp called Sun-Lite Model D. It is claimed 
that this lamp gives steady and uniform screen illumination in spite 
of the sensitiveness of the Supro copper-coated carbons employed. 

The Projection Practice Committee of the S. M. P. E. has proposed 
the use of glass mirror guards as a means of preventing loss of light 
due to reflector pitting in reflector arc lamps. 67 

For 35-mm. projection in small theaters and auditoriums there 



26 REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 

has recently been developed by the General Electric Company a 
2 100- watt, 60- volt projection lamp, containing a bi-plane filament 
and the new bi-post base. This lamp gives the highest screen 
illumination attainable with any filament lamp in the 35-mm. 
projector and gives more than double that of the heretofore generally 
used 900- watt, 30-ampere projection lamp. 

With rims and spokes of clock-spring steel, DeVry has manufac- 
tured a de luxe film reel, which will not remain bent under pressure 
or blows, and consequently saves the life of film by its permanent, 
true alignment. The reel holds 2000 feet of film. A novel feature 
is the sliding attachment of the spokes to the hub which allows 
expansion under strain and prevents warping, dishing, etc., common 
with the usual reel. Automatic clipping is accomplished by wells in 
the hub. The operator merely presses the film slightly over the well 
with the finger. Prongs catch the perforations automatically with 
a non-slip grip. 

From Germany comes announcement that the Zeiss Ikon A. G. 
have made various valuable improvements to their projectors called 
Ernemann IV, Ernon IV T, and Ernemann VII. Especially inter- 
esting is that for theater machines : film reels have been issued which 
take 1300 meters of film. These reels are now available for all the 
projectors made by Zeiss Ikon. Furthermore, all their apparatus is 
now fitted with a so-called "protector" arrangement, which is a 
fire protection arrangement cutting off the light-beam when the film 
stands still, although the machine continues to run. All the ma- 
chines are fitted with a rotating shutter between the light-source 
and the gate, and are obtainable for right-hand and left-hand han- 
dling. It should be mentioned that all the Zeiss Ikon projectors use 
the new Zeiss Ikon lamps. 

The Klangfilm Company have also improved their apparatus. 
There will also be found a description of the new Europa sound- 
head built by Klangfilm, which is fitted with a rotating sound-gate. 68 

The well-known projectors of the firm of Eugen Bauer G. m. b. H., 
Stuttgart, namely, Super 7, Standard 7, and Standard 5, are now fitted 
with the so-called Bauer Doppel-Ausgleichs-Gerat. During the last 
year this "Doppel-Ausgleichs-Gerat" was tried very successfully in 
several large theaters. It is said to insure a great security of opera- 
tion and a very high quality of sound reproduction. The main fea- 
tures of the construction are the special type of driving mechanism, 
the possibility of incorporating it in the projector, the use of a double- 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 



27 



way clutch and a special micro lens with optical slit, and finally, a 
highly effective photo-cell. 

(IV) FILM LABORATORY PRACTICE 

It is unfortunate that we are unable to report many items of prog- 
ress either in laboratory equipment or technic. The release print 
situation in this country remains 
decidedly spotty. It is quite 
apparent that producers do not 
take the release print seriously. 
This is a regrettable situation 
because, after all, the public sees 
only these prints. Further educa- 
tional work to impress producers 
with the necessity of giving the 
release print more attention is 
apparently in order. 

In the line of laboratory equip- 
ment, Metro - Goldwyn - Mayer 
Studio reports the use of a tur- 
bulator in their sound and picture 
negative developing machines. 
The use of this turbulator prac- 
tically eliminates the so-called ''di- 
rectional effect" and appears to 
produce a picture and sound 
negative of distinctly improved 

quality. With this turbulator in use, H&D strips sent through 
"heads first" are indistinguishable from those sent through "tails 
first." 

(V) APPLICATIONS OF MOTION PICTURES 

(1) Education. The growth of sound motion pictures in educa- 
tional work in the United States has been hampered by the lack of 
suitable films and by the failure of the equipment makers to solve 
the problem of providing sound-film reproducers at a price that the 
average school can afford. In Europe considerable progress is 
reported in the use of 16-mm. sound pictures in the public schools. 
The movement is strong in Germany, where the Government is 
sponsoring a campaign of propaganda in the schools to further the 




(Courtesy of Electrical Research Products, Inc.) 

FIG. 13. Horse race finish at Santa 

Anita track. 



28 REPORT OF PROGRESS COMMITTEE [J. S. M. P. E. 

aims of the Administration. Sweden also reports progress in the use 
of motion pictures as an adjunct to elementary education. 

(2) Race Timing Devices. The Electrical Research Products, 
Inc., in cooperation with the Eastman Kodak Company, have 
developed and introduced into the horse racing sport an electrically 
controlled precision timing and judging system. The equipment 
photographs the time of predetermined intervals of the race and 
also photographs the finish and the total elapsed time (Fig. 13). It 
consists of a high-speed 16-mm. camera, placed at the finish line 
and directly connected to a rapid film processing and enlarging 
equipment. Integral with the camera is an electrically controlled 
clock which is started by an interruption of the light-beam focused 
upon the photoelectric cell situated at the start of the race. Interval 
times are photographed by similar successive interruptions of the 
light-beams focused upon the photoelectric cells at other points along 
the track. The equipment can be operated to deliver in less than 
three minutes after the finish of a race a print of the finishing frame 
showing the elapsed time and the order of the finish. 

(VI) PUBLICATIONS AND NEW BOOKS 

With few exceptions, the periodicals of interest to the readers of 
motion picture literature have been continued. In the field of visual 
education, an interesting publication was overlooked when compiling 
the 1932 report. This is the journal, Sight and Sound (London), 
which was initiated in 1932. A small publication bearing the name 
Journal of the Motion Picture Society of India (Bombay) made its 
first appearance in January of this year as the official organ of the 
Motion Picture Society of India. A new periodical known as Ama- 
teur Cine World was added to the list of amateur publications. Per- 
sonal Movies (Canton, Ohio) was discontinued at the end of 1934. 

A list of the principal books that have been published since the 
last report of the Committee (April 1934) follows: 

(1) Year Book of Motion Pictures (1935), 16th Edition; Film Daily, New 
York, N. Y. 

(2) Motion Picture Almanac (1934); Quigley Publishing Co., New York, 
N. Y. 

(3) Kinematograph Year Book (1935); Kinematograph Publications, Ltd., 
London. 

(4) Abridged Scientific Publications of the Kodak Research Laboratories 
(Vol. XV); Eastman Kodak Co., Rochester. 

(5) Publications from the Scientific Laboratory, Agfa Photographic Division 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 29 

(Veroffentlichungen des wissenschaftlichen Zentral Laboratoriums, 
Agfa Photographischen Abteilung) (Vol. Ill); /. G. Farbenindustrie 
Aktiengesellschaft, Hirzel, Leipzig. 

(6) Yearbook of the Cine-Amateur (Jahrbuch des Kino-Amateurs 1935), 
edited by W. Frerk; Photokino Verlag., Berlin. 

(7) The Photographic Darkroom; E. J. Wall, American Photographic Pub 
lishing Co., Boston. 

(8) Photographic Technic (La Technique Photographique), 2nd Edition; 
L. P. Clerc, Montel, Paris. 

(9) Chemistry of Developers and Development (in Russian); V. I. Shiber- 
stoff, Government Printing Office for Light Industry, Moscow. 

(10) Cinematographic Technic (La Technique Cinematographique), 4th 
Edition; L. Lobel and M. Dubois; Dunod, Paris. 

(11) Filming with the Cine Kodak Eight (Filmen mit Cine Kodak Acht); 
A. Stuler, Knapp, Halle. 

(12) Physics of Sound Films (Physik des Tonfilms); A. Haas; Teubner, 
Leipzig. 

(13) Sound Motion Pictures; J. R. Cameron, Cameron Publishing Co., Wood- 
mont, Conn. 

(14) Modern Acoustics; A. H. Davis, Macmillan Co., New York. 

(15) Applied Acoustics; H. F. Olson and F. Massa, P. Blakiston's Sons Of Co., 
Inc., Phila., Pa. 

(16) Home Processing; P. W. Harris, Geo. Newnes, Ltd., London. 

(17) Making Home Movies; D. C. Ottley, Geo. Newnes, Ltd., London. 

(18 Projection Room Regulations and Practices; R. Ruedy, National Re- 
search Council, Ottawa, Canada. 

(19) Motion Pictures in Education in the United States; C. M. Koon, 
University of Chicago Press, Chicago. 

(20) Two New Systems of Cinematography in Relief (Due Nuovi Sistemi di 
Cinematographia in Relievo); G. Jellinek, Libreria Editrice Politechnica, 
Milan. 

(21) The Physics of Electron Tubes; L. R. Koller, McGraw-Hill Book Co., 
New York. 

APPENDIX A 

General Field of Progress of the Motion Picture Industry in Great 

Britain 

General. For the British Motion Picture Industry, the year 1934 
has been a year of progress. The output of British films has been 
high, nearly 300 having been registered, including about 200 of 3000 
or more feet in length. Theater box-office receipts have reflected 
the improved industrial situation prevailing throughout the country, 
and new theaters continue to open at a steady rate of about 100 an- 
nually. There have also been extensive and significant develop- 
ments in the home entertainment, educational, and industrial fields. 



30 REPORT OF PROGRESS COMMITTEE [J. S. M. p. E. 

Producers and Studios. During the year, a number of new pro- 
ducers have appeared, and existing producers have extended the 
premises. The Blattner Studios, Elstree, have been acquired by 
Messrs. Joe Rock and Leslie Fuller, and equipped with a "Visa tone" 
sound recording system. Interworld Studios, Isleworth, have been 
opened as service studios, with Western Electric sound. Additional 
stages have been constructed at Twickenham Film Studios, with 
"Visatone" sound, and Associated Talking Pictures (RCA High 
Fidelity Sound). London Film Productions rented various sound 
and silent stages during the year, including one at British & Do- 
minions Studios, Elstree, who have now extended their lot by ten 
acres. London Film Productions are Western Electric licensees. 
Warner Bros.-First National Productions have also built fresh pre- 
view and cutting rooms, workshops, etc., the exterior of the new build- 
ings providing suitable settings for a variety of outdoor shots. This 
producer has also become a Western Electric licensee. The Welwyn 
Studios, associated with British International Pictures, have been 
equipped with an Ambiphone sound recording system. 

Photography and Film Laboratories. Much interest has been shown 
in color photography and projection during the year. One studio 
produced a picture (Radio Parade), in which certain sequences were 
photographed using the Spicer-Dufay color process; another pro- 
ducer is experimenting with the Hillman process and a laboratory 
has been installed for the processing of industrial cartoons produced 
by the Gasparcolor Process. Spicer-Dufay are also experimenting 
with 16-mm. film, and tests of this system for newsreel work have been 
made. Apart from those required by color photography, no techni- 
cal changes in film processing of major importance appear to have 
been seriously considered. No new types of motion picture film were 
marketed during the year, but the problem of loss of sound in printing, 
aggravated by the extended high-frequency ranges now recorded, has 
stimulated research into the cause of printer slippage and its possible 
cure by reducing the perforation pitch of sound negative material. 
Paramount Sound News have brought anti-halation stock into use, 
to increase the high frequencies. 

The larger laboratories have given further attention to the control 
of both positive and negative developers by sensitometric means, and 
at the end of the year three laboratories were equipped with the 
Eastman Type 2-B Sensitometer. Over the year, the average gamma 
to which negative picture was developed was 0.65, while positives were 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 31 

developed to 2.10-2.40. One laboratory has installed new develop- 
ing equipment made at its own works and capable of operating satis- 
factorily at a speed of 180-200 feet a minute. Messrs. Watson & 
Sons, London, have developed a photoelectric densitometer with 
automatic curve-plotting attachment, for use with an Eastman sensi- 
tometer, and from Germany comes the announcement of a new 
photoelectric microdensitomer with alternative photoelectric or 
photographic recording. 

The local manufacture of cameras and associated equipment con- 
tinues to expand, among the most striking products during the year 
being the Skyhi camera cranes. These cranes are pneumatically 
driven, and will rise to a height of 20 feet from the ground. They 
may be made to rise at speeds up to 200 feet a minute, and a setting 
mechanism is provided to enable any rate of rise to be repeated auto- 
matically. In each crane the boom and its platform can be rotated 
through 360 degrees, and the forward platform through 180 degrees. 
The weight of a crane is about 3 x /4 tons. It is completely silent in 
operation. 

Sound Recording Equipment. During the year, in addition to the 
developments referred to above, other new and improved sound re- 
cording equipment has been produced. British Acoustic Films, 
Ltd., have introduced a "Full Range" recording equipment operating 
from a-c. mains, with "noiseless recording." This firm has also pro- 
duced an improved light-weight combined sound and picture camera, 
and a new ribbon microphone. 

The Phillips-Miller recording system has been on trial at Elstree 
and at the British Broadcasting Corporation's laboratories in London. 
In this system, a recording sapphire inscribes a variable-width track 
in a black coating carried on a 17.5-mm. film. The track can be 
used for reproduction immediately after recording, and can be printed 
without any processing. 

In the newsreel field, both Gaumont News and Pa the* News have 
acquired Vinten cameras, and Gaumont News have also adopted 
British Acoustic sound. In view of the restriction of the number of 
transmitters imposed at important public ceremonies, the British 
newsreel companies are finding it convenient to use common input 
channels for their main sound recording amplifiers. 

Exhibitors and Theaters. Among the exhibitors in England, the 
year generally has been characterized by some increase in returns, as 
already mentioned, but the trade has also been faced with an increase 



32 REPORT OF PROGRESS COMMITTEE [J. S. M. P. E. 

of competition, due to new cinemas and to free propaganda and ad- 
vertising shows. Overbuilding principally affects those exhibitors 
who have failed to keep pace with the taste of the public. The 
churches are beginning to become film minded, and in some districts, 
developments due to this and to "clean film" campaigns are causing 
concern. In view of this competition the appearance of a number of 
films of high artistic merit and unquestionable taste has been oppor- 
tune, and such films appear to appeal to a large public. 

Technically, there have been no very great developments in the 
standard of projection, but the leading exhibitors appear to be keep- 
ing in mind the possibility of the introduction of wide film in the next 
few years. During the year, new projectors have been introduced 
by British Acoustic Films and the British Thomson-Houston Co. 

On the acoustic side, the principal feature has been a steady im- 
provement and increase of the frequency range in sound reproduction. 
The Western Electric Company reports a steady stream of conver- 
sions to Wide Range, and toward the end of the year this company 
announced that all future installations would normally be Wide 
Range. In addition, all future systems will be all-mains, thus elimi- 
nating the need for batteries. RCA High Fidelity reproducing equip- 
ment has been improved to give better frequency and acoustic re- 
sponse, and both British Acoustic and British Thomson-Houston 
systems have been adapted to cover the extended frequency ranges 
now demanded in reproduction. 

Industrial and Educational. During the year, there has been a 
considerable increase in the number and quality of 16-mm. sound- 
film equipments available, and organizations exist for servicing and 
providing shows at moderate charges in any part of the country. 
One large tobacco firm financed eight thousand propaganda shows, 
and the Electrical Development Association has now commenced a 
campaign with specially produced films of high merit to sell electric- 
ity. The production of educational films has commenced on a fairly 
large scale, the Kodak Company being well to the fore with an im- 
portant collection of medical films. There is some confusion in this 
field, due to the existence of three alternative standards, the D. I. N., 
the S. M. P. E., and the Pathd. 

Amateur and Commercial. The Kodak Company reports that the 
most important advance during the year for amateurs has been the 
improvement and cheapening of artificial illuminants, affording the 
amateur an immensely increased range of activity at reasonable cost. 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 33 

In this connection, the General Electric Company, London, have come 
into prominence with their Photoflood lamp. Moreover, together 
with the illumination advances, faster lenses have been introduced 
for 16-mm. and 8-mm. work, and having regard to the possibilities 
of supersensitive panchromatic film and the use of filters, the pros- 
pects of development in the use of cinematography by amateurs and 
professional portrait or commercial photographers are very bright. 
For commercial photography, the ultra-high-speed Eastman camera 
with Western Electric timing device is finding many important ap- 
plications. 

Broadcasting. Radio broadcasting during 1934 has both contrib- 
uted to and derived much from the stock-in-trade of the motion pic- 
ture industry. Gaumont British arranged for the radio transmission 
of a newsreel film of the finish of the England- Australia race. The 
television report of the Selsdon Committee foreshadows a new source 
of competition to exhibitors. Film subjects, such as Walt Disney's 
famous productions, contribute to the delights of an evening radio 
program; radio features form attractive subjects for film producers, 
and so on. 

Altogether it would seem that although the past year has been 
mainly one of consolidation for the industry, the future presents the 
possibility of several very important advances. 

APPENDIX B 

General Field of Progress of the Motion Picture Industry in Japan 

The chief development in the motion picture industry of Japan 
during 1934 was increased activity in sound production. The total 
footage of pictures produced was between 5 and 10 per cent lower than 
in 1933, but the recorded footage was very much higher. The pro- 
ducers are finding it increasingly difficult to market silent pictures, 
the exhibitors asking that the picture be at least scored. Forty 
per cent of the sound pictures produced in 1934 were scored. These 
scored pictures offer poor competition to an "all talkie" or synchro- 
nized picture, however, as indicated by the 10 most popular pictures 
for 1934: 5 all talkie; 1 part talkie; 1 sound (scored); 3 silent. 

Theater returns were exceptionally good during the first half of 
1933 but dropped off considerably during the latter half, due partly 
to the disastrous typhoon that occurred in the summer. The studios 
lost a number of stages, interfering with production. Also, the pur- 



34 REPORT OF PROGRESS COMMITTEE [j. s. M. p. E. 

chasing power was noticeably reduced in the areas hit by the storm. 

Several new and modern theaters were opened in Tokyo in 1934 
and one in Osaka. These theaters are showing mainly imported 
pictures. Last year there were more pictures imported than any 
year since the advent of sound, about 275 from the United States 
and 50 from Europe. European pictures are quite popular in Japan. 
In a list of 16 best pictures put out by the Educational Department 
there were 6 from the United States, 5 from Germany, 2 from France, 
and 3 from Japan. Of the 10 most successful pictures as listed by 
Movie Times, 8 were produced in the United States, one was assembled 
and scored locally by Fox Movietone, and one was imported from 
France. 

Among local productions, so-called modern pictures continued to 
gain over the costume plays which were not so long ago the main 
source of material for the movies. 

With the coming of sound there has been quite an increase in the 
number of modern cameras in use. There has also been a number 
of better class stages erected for the purpose of producing sound pic- 
tures. In the laboratories the use of continuous machinery has in- 
creased and other laboratories are under construction which will use 
nothing but continuous equipment. As these modern equipments 
and methods have come more into use and the technicians have be- 
come better acquainted with sound, the general technical quality of 
the pictures has showed a marked improvement. 

Sixteen-mm. pictures have come into more extensive use for edu- 
cational and propaganda purposes. At present there is very little 
16-mm. sound, but several laboratories are experimenting upon re- 
ducing 35-mm. sound pictures to 16-mm. 

REFERENCES 

1 J. Soc. Mot. Pict. Eng., XXIII (Sept., 1934), p. 160. 

2 Camera (Luzern), 13 (July, 1934), p. 3. 

3 Photo Revue, 46 (1934), p. 373. 

4 Kinotechnik, 17 (1934), No. 6, p. 98. 

6 Ibid., 17 (1934), No. 11, p. 175. 
Ibid., 17 (1934), No. 8, p. 127. 

7 Ibid., 17 (1934), No. 16, p. 255. 

8 Ibid., 17 (1934), No. 19, p. 308. 

9 Kino-Photo Institute (Moscow), 2 (1935). 

10 Ibid. 

11 J. Phys. Chem., 38 (June, 1934), p. 817. 

12 J. Opt. Soc. Amer., 25 (Jan., 1934), No. 1, p. 4. 



July, 1935] PROGRESS IN MOTION PICTURE INDUSTRY 35 

13 Kino-Photo Institute (Moscow), 2 (1935), p. 19. 
"Ibid., 2 (1934), p. 30. 

15 Econ. Rev. Soviet Union, 10 (June. 1935), p. 8. 

16 Mot. Pict. Dev. U. S. Dept. Commerce Reports, April 12, 1934. 

17 U. S. Patents 1,939,171, 1,954,337; Ger. Patent 579,078. 

18 Brit. Patents 385,111, 386,442. 

19 Amer. Cinemat., 14 (April, 1934), p. 492. 

20 Ibid., 13 (Dec., 1933), p. 312. 

21 U. S. Patents 1,939,058, 1,947,669, 1,950,166, 1,954,340, 1,955,590, 1,955,591, 

1,955,617, 1,967,214, 1,967,215, 1,967,836; Brit. Patents 381,662, 386,- 
517, 397,261, 407,156. 

22 /. Opt. Soc. Amer., 24 (March, 1934), p. 73. 

23 Amer. Cinemat., 15 (Oct., 1934), p. 248. 

24 Ibid., 15 (July, 1934), p. 130. 
26 Brit. Patent 411,748. 

26 /. Soc. Mot. Pict. Eng., XXIV (Feb., 1935), No. 2, p. 110. 

27 Ibid., XXIV (June, 1935), No. 6, p. 487. 

28 Brit. J. Color Supp., 27 (Dec. 7, 1934), p. 47. 

29 Fortune, 10 (Nov., 1934), p. 88. 

30 Ibid., 10 (Oct., 1934), p. 92. 

31 Brit. J. Color Supp., 27 (Aug. 3, 1934), p. 30. 

32 Mot. Pict. Herald (Feb. 16, 1935). 

33 Home Movies and Home Talkies, 13 (1934), No. 4, p. 137. 

34 Filmtechnik, 18 (1934), p. 8. 

36 Kinoamateur (Dec. 1, 1934), No. 12. 

36 Film fur alle, 8 (1934), pp. 9, 268. 

37 Ibid., 8 (1934), pp. 12, 366. 

38 Br. J. Phot. (Jan. 25, 1935), p. 50. 

39 Filmtechnik, 18 (1934), No. 8, p. 201. Amer. Cinemat. (Nov., 1934), p. 326. 

40 Amer. Cinemat. (Nov., 1934), p. 326. Camera (Jan., 1935). Internal. 

Phot. (Feb., 1935). 

41 Movie Makers (Feb., 1935), p. 78. 

42 Amer. Cinemat. (Sept., 1934). 

43 Kinoamateur (Dec. 1, 1934), No. 12. 

44 Ibid. 

46 Camera (Jan. 7, 1934). Phot. J. (Jan., 1935), p. 28. 

46 J. Soc. Mot. Pict. Eng., XXHI (July, 1934), No. 1, p. 14. 

47 Photofreund, 14 (Aug., 1934), p. 279. 

48 Filmtechnik, 18 (1934), p. 8. 

49 J. Soc. Mot. Pict. Eng., XXIH (Aug., 1934), No. 2, p. 87. 
60 Movie Makers (Feb., 1935), p. 78. 

51 Amer. Cinemat. (March, 1935), p. 114. /. Soc. Mot. Pict. Eng., XXIH 

(Aug., 1934), No. 2, pp. 96, 108. Ibid., XXTV (Jan., 1935), No. 1, p. 63. 
Ibid., XXIV (Feb., 1935), No. 2, p. 95. Phot. J. (June, 1934), p. 471. 

52 /. Soc. Mot. Pict. Eng., XXIII (Aug., 1934), No. 2, p. 96. 
63 Movie Makers (Jan., 1935), p. 10. 

"Phot. J. (June, 1934). 

66 Kinotechnik, 15 (March, 1934), p. 86. 



36 REPORT OF PROGRESS COMMITTEE 

66 Amer. Cinemat. (March, 1935), p. 114. 

67 Ibid. (Nov., 1934), p. 326. 

68 Movie Makers (Jan., 1935), p. 10. 

69 Ibid. (Feb., 1935), p. 91. 

60 La Cinemat. Franc., 16 (1934), p. 830. Photohaendler (July 24, 1934), p. 458. 

61 Film fur alle, 8 (1934), No. 12, p. 366. 

62 Photowoche, 24 (1934), No. 13, p. 623. 

63 Ibid. 

64 Filmtechnik, 10 (1934), No. 2, p. 23. 

66 Home Movies and Home Talkies, 3 (1934), No. 6, p. 232. 

66 Electronics (Feb., 1935), p. 52. 

67 /. Soc. Mot. Pict. Eng., XXIII (Dec., 1934), No. 6, p. 365. 

68 Filmtechnik, 20 (1934), p. 237. Ibid., 4 (1935), p. 42. Kinotechnik, 8 

(1935), p. 135. 



TELEVISION AND MOTION PICTURES* 
ALFRED N. GOLDSMITH** 



Summary. Ultra-short-wave television transmissions with pick-ups using 
electron scanning, together with cathode-ray tube reception, are regarded as likely 
to be acceptable for television broadcasting to the public. Comparison of the received 
television pictures with theater and home motion pictures indicates superiority of 
the latter but shows that satisfactory entertainment should also be obtainable from the 
former. For various physical and psychological reasons which are listed, the vogue 
of the theater is likely to continue, provided motion picture methods are constantly 
kept up-to-date and that television technic is studied and utilized to such an extent 
as may prove practicable. 

It must be admitted that the relationship between motion pictures 
and television has, in the past few years, been made the subject of 
numerous effusions which, even from a charitable point of view, must 
be characterized as highly imaginative and distinctly misleading. It 
is unfortunate that the present and future correlation of these im- 
portant fields should have been made the subject of casual publicity 
releases or of selfishly inspired propaganda. The subject is of con- 
siderable importance and fully merits thoughtful and impartial 
analysis. Such analysis requires, on the other hand, an unusually 
complete knowledge of the commercial activities and engineering 
methods of the two fields which are involved and perhaps something 
of a gift of prophecy as well. For these reasons, the task of presenting 
a necessarily brief and summarized analysis of the relations between 
motion pictures and television must be approached with some diffi- 
dence and hesitation. Of necessity, numerous important factors 
must be omitted and others must be comparatively slighted within 
the limited space of such a presentation as this one. 

It is, nevertheless, proposed in the following : 

(a) to consider the methods likely to be used in television-telephone broad- 
casting into the home; 

(6) to compare the results likely to be achieved by television-telephone broad- 
casts into the home with the results attainable by theatrical sound motion pictures; 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** New York, N. Y. 

37 



38 A. N. GOLDSMITH [j. S. M. P. E. 

(c) to consider the probable points of contacts between these fields, and to 
explore the possibilities of cooperative effort between them; and 

(d) to consider the possible general effects of the wide-spread acceptance of 
television-telephone broadcasting into the home on the motion picture theater. 

In studying a, the first of these topics, it is necessary to adopt some 
standard for "television" a term having widely different meanings 
to various people. It is proposed to accept "standards" which repre- 
sent what may reasonably be expected to be attained on a large scale 
within the next few years provided mass production of equipment for 
television-telephone broadcasting reception in the home is carried 
out. The term "tele vision- telephone broadcasting" is used because 
it is naturally assumed that the television picture will be consistently 
accompanied by the corresponding sound or telephonic material. 
While some of the details of home television-telephone reception will 
be given below under the study of topic b, it may be postulated here 
that such television will be accomplished by the use of 

(*) an electronic pick-up rather than a mechanico-optical pick-up. (Such 
pick-ups include the iconoscope and the dissector tube.) 

(ii) an ultra-short wave transmitter or transmitters for the television and 
telephone portions of the program. 

(Hi) a coaxial-conductor cable, or its equivalent, or an ultra-short-wave radio 
relay system, for the syndication of the program material for network operation; 
and 

(iv) an electronic receiver of the cathode-ray type, with a fluorescent image 
screen, rather than a mechanico-optical receiving system. It is impracticable 
within the limits of this paper to discuss the principles, design, construction, or 
operation of the complicated devices mentioned above. 

Proceeding to the study of topic b, there will be given an itemized 
and instructive comparison of the practical results to be expected by 
home television-telephone reception as compared with motion picture 
theater performance. 

(1) Mode of Picture Production. The methods of producing the 
pictures are entirely different in the two cases, odd as that may seem. 
The theater picture is projected as a complete unit, one entire frame 
at a time. The delineation is produced and limited by aggregates 
of silver grains in the developed positive image. The television pic- 
ture is produced by a luminous dot (or "dot-element"), the brightness 
of which is accurately controlled as it passes in succession over a 
series of parallel and closely adjacent lines until it has covered the 
entire area of one frame. In the theater case, the entire picture is on 
the screen at the same time, to be succeeded by darkness prior to the 



July, 1935] TELEVISION AND MOTION PICTURES 39 

projection of the next frame, and so on. In the television case, there 
is never anything more than a more or less bright dot upon the screen! 
The television picture depends even more upon persistence of vision 
than the theater picture, being, in fact, nothing more than a flickering 
and flying dot. It must be added that certain technical details of 
television picture production have not as yet been standardized. 
However, the above-cited features appear likely to be permanently 
present. 

(2) Number of Picture Elements. The number of picture elements 
determines the detail or (roughly) the story-telling capabilities of the 
picture. In round numbers, the theater picture has something of the 
order of 5,000,000 picture elements, whereas even a good home tele- 
vision picture will probably have something like 150,000 elements. 
This is a ratio of 30 to 1 in favor of the theater picture. However, 
it must be noted that the entertainment value of a picture in motion 
(whether produced by projection or by television) is not in direct 
proportion to the number of picture elements which it contains, so 
that we are not entitled to draw the conclusion that theater pictures, 
though more detailed in structure, are necessarily far more enter- 
taining (particularly on the small home screen) than television pic- 
tures. Probably a television picture in the home will be described 
by most as a "fair home movie." 

(3) Grain or Line Structure. Theater pictures of reasonable size 
from a suitable positive show negligible grain if viewed at moderate 
and practicable distances, and of course show no line structure (for 
monochrome pictures). Television pictures show no grain structure, 
but may show a slight line structure if viewed too closely. However, 
high-detail television pictures, viewed at normally comfortable dis- 
tances, will show practically no line structure and certainly no 
objectionable line structure. 

(4) Color of the Picture. Theater pictures are normally black in 
the shadows and white (blue-white or yellow-white) in the highlights. 
When projected from toned or tinted positives, they show the corre- 
sponding hue. Television pictures are also practically black in the 
shadows, but the highlights may be bright yellow, greenish yellow, or 
even a practically neutral white. The latter color will probably 
become common practice in television as development of the art 
proceeds. 

(5) Possibility of Full-Color Pictures. It is readily possible today 
to produce theater pictures which show substantially the colors of 



40 A. N. GOLDSMITH [J. S. M. P. E. 

nature or at least an acceptable approximation thereto, although 
there are definite economic handicaps in production and reproduction 
of such pictures. Television in full colors seems to be an almost 
impracticable proposition in the present or likely early state of that 
art, although small-scale demonstrations of its abstract possibility 
have indeed been given. 

(6) Size of the Picture. Theater pictures range in size from, say, 
6 by 8 feet to perhaps 18 by 24 feet, or even more in special cases. 
Thus, their area is between 48 and 32 square feet. Home television 
pictures range from about 6 by 8 inches to perhaps 18 by 24 inches 
or, in special cases, somewhat more (though generally at the cost of 
picture detail and brightness). Thus their area lies between about 
0.3 and 3 square feet. On this basis the area of the theater picture 
is about 100-160 times that of the home television picture. A more 
normal comparison would be with the approximate 30 by 40-inch 
home motion picture, having an area of about 8 square feet or, say, 
about 5 times that of the average television picture. 

(7) Picture Brightness. Theater pictures are generally ade- 
quately bright for viewing in a darkened auditorium (that is, an audi- 
torium with illumination about 0.5 foot-candle). The television 
pictures are also sufficiently bright to be viewed in a dimly lighted 
room but dark window shades will be required for daylight hours, 
and for the evening as well if the street lighting outside the home is 
at all bright. 

(8) Flicker of the Picture. The theater picture consists of 24 
frames per second, each of which is generally projected twice before 
the next frame reaches the screen. Flicker is absent, although traces 
of an effect depending upon picture sequence are still found in the 
case of rapidly moving objects and in the stroboscopic backward- 
turning of the wheels of pictured vehicles. Television pictures may 
be projected in two sets of 30 pictures each, the two sets being pro- 
jected in 1 second. Interlaced scanning may be used, and under 
these conditions a substantially flickerless picture is obtained. De- 
spite the projection of 60 half-detail pictures per second by this 
method (equivalent closely to 30 full-detail pictures per second), it is 
possible to use ordinary 24-frame-per-second motion picture film for 
the television subject without undue difficulty by the use of technical 
expedients which can not be here described. 

(9) Viewing Distance. Taking an optimum viewing distance of 
4.5 or 5 times the picture diagonal, theater pictures may be most 



July, 1935] TELEVISION AND MOTION PICTURES 41 

conveniently viewed from 45 to 135 feet from the screen, while home 
television pictures will be viewed from about 4 to 11 feet from the 
screen. This is a ratio of viewing distances of about 11 to 1 in the 
two cases. 

(10) Audience Size. Long experience has demonstrated that the 
comfortable size for theater audiences ranges from 500 to 5000 per- 
sons, with perhaps some doubt at one extreme or the other. The 
corresponding home audience may be expected to include from 3 to 
15 persons, a ratio in favor of the theater of about 200 to 1. It must 
not be inferred, however, that the economic ratio for the two fields is 
anything like as high as this indeed it has not yet been determined 
just what will be the cost per person per hour of entertainment for 
home television-telephone broadcasting. 

(11) Synchronism of Picture with Sound. In the theater, the pic- 
ture and sound are correctly associated within x /24th of a second, 
assuming proper editing and threading. In the case of home tele- 
vision-telephone programs, the synchronism is even closer (though 
this is not noticeable as an advantage), and is entirely correct and 
automatic. Some rather romantic writers on the subject have dilated 
upon the "marvel" of the synchronism of picture and sound in such 
programs. As a matter of fact, considering the fundamentals of the 
processes employed, it would be even more marvelous if synchronism 
were not attained for tele vision- telephone broadcasting reception. 

It is not practicable at this time, before mass production of tele- 
vision equipment has been initiated, to give a reliable comparison of 
the cost of theater and home equipment. In a general way it may 
be said that theater equipment costs in the thousands of dollars and 
home equipment about the same number of hundreds of dollars, thus 
giving a cost ratio of perhaps 10 to 1. Here again some caution 
must be used in interpreting such figures since there are numerous 
other economic factors involved in a valid comparison. 

While it is not feasible within the limits of this presentation to 
give even an outline of the various methods employed in modern 
television, some numerical data concerned with picture detail may be 
included as of present interest. These consist, first, of a personal 
opinion, in motion picture terminology, of the value and character- 
istics of television pictures having various numbers of dot-elements 
composing them. The figures are understood to be merely generally 
descriptive, but it is believed they are instructive in judging the "mo- 
tion picture value" of various television systems: 



42 A. N. GOLDSMITH [j. S. M. P. E. 

Elements in Picture "Motion Picture Value" 

10,000 A fair close-up of one person (head and shoulders). 

20,000 Two persons in a moderately good close-up (though without 

fine detail). 

40,000 Fair medium shots. 

80,000 Good medium shots and fair long shots. 

160,000 Excellent close-ups of several persons, good medium shots, 

and acceptable long shots (except for unusual "pageant" 
subjects and the like). 

Taking the last-mentioned type of television picture, and assuming 
flickerless transmission, it is found that the required "side-bands" 
produced by the picture modulation of the ultra-short-wave carrier 
have a width of the order of 1.5 megacycles (or about 150 times the 
frequency band required for high-fidelity 10,000-cycle sound repro- 
duction) ! 

Passing to topic c above, namely, the contacts and cooperative 
possibilities between motion pictures and television, it is clear from 
the beginning that there can be a close connection if such is desired. 
A person viewing a small picture in motion with synchronized sound 
might find some difficulty in knowing whether he was viewing a sound 
motion picture projected from film or a television-telephone broad- 
casting reception. He might be even more puzzled if the subject 
matter were, say, a newsreel used to control the television-telephone 
transmitter, an entirely feasible procedure. Obviously the technic 
of producing a television-telephone broadcast program will closely 
resemble that of producing a sound motion picture. Methods of 
costuming, make-up, script construction, "camera" technic, sound 
pick-up, set construction and illumination, and the like may well be 
similar in the two fields, though probably not with the same degree 
of elaborateness in the case of television. There is one respect in 
which they will necessarily differ if an original performance (rather 
than a film record) is broadcast. This is a limitation of television- 
telephone broadcasting, namely, the possibility of only one "take," 
viz., the one that is broadcast. In motion picture production, any 
reasonable number of takes may be made; not so in broadcasting, 
where the radio wave irrevocably carries the selected performance to 
all homes. 

As has been mentioned above, sound motion picture films may be 
excellent subject matter for programs from some stations, and may 
even afford one means of syndicating programs in somewhat the same 
way electrical transcriptions (phonograph disk records of programs) 



July, 1935] TELEVISION AND MOTION PICTURES 43 

are now used. It is not believed, however, that television-telephone 
syndication will be fully satisfactory unless there are also actual inter- 
connecting wire or radio networks between the outlet stations, since 
there will be many occasions (for example, a speech by the President, 
a political convention, an evening prize fight, and the like) when 
the public can hardly be completely satisfied by any radio perform- 
ance which does not take place at the same time as the actual event. 
Indeed it must be admitted that this is one of the outstanding capa- 
bilities of radio broadcasting which it would be unwise to discard. 

Many persons are convinced that television broadcasting will whet 
the appetite of the "lookers," and, so far from diminishing the theater 
audience, will build it up by arousing interest among children and 
adults alike in the probably more elaborate and highly developed 
offerings of the theater. It is also clear that the theater can, to a 
considerable extent, utilize radio advertising by television-telephony, 
for example, by the sponsored transmission of trailers of one sort of 
another. Radio will then offer the theater a remarkably effective 
method of submitting its "sample line" to the public. 

This brings us to topic d above, namely, the possible effect upon 
the theater of the wide-spread acceptance of television-telephone 
broadcasting. We are inclined to be definitely optimistic as to this. 
The argument that television broadcasting may keep people out of 
the theater does not appear to have much weight. Consider, for 
example, the following controlling principles: 

(1) Intrinsically the home is certainly not so good a showplace as 
the theater. It is more difficult to suppress natural and man-made 
noises in the home; here manners tend to be more "free and easy" 
than is desirable for showman-like presentations; the problem of 
setting up the theater in the home is far from simple when furniture 
must be moved to afford a good view of the screen and the home folks 
and guests located in the corresponding convenient viewing positions ; 
and home lighting is rarely as controllable or suitable for picture pres- 
entation as is the case in the theater. Indeed, the customary sur- 
roundings of the home are not especially favorable for the creation of 
a world of illusion, which has always been the successful function of 
the theater. It is not maintained that there will not be value and 
interest to the home presentation quite the contrary. It is stressed, 
however, that the home has certain disadvantages of long standing 
for program presentation which can not be disregarded. 

(2) Conversely, the theater has a number of definite and inherent 



44 A. N. GOLDSMITH [j. s. M. p. E. 

advantages as a showplace. It arouses the interest of the audience 
by heavy theater advertising in the press, by the play-up of the "fan 
magazines," and by other exploitation methods known to skillful 
managers, thus creating in the prospective audience the proper mood 
of pleasurable anticipation. The marquee and lobby of the theater, 
ablaze with light and motion, and with attractive photographs of 
selected scenes from the picture displayed within, further attract the 
audience. Within the theater, suave but real discipline is maintained 
by the ushers a task calculated to daunt the bravest in the home. 
Furthermore, the price of admission, exacted at the box-office just 
before entry, is a powerful deterrent to lack of interest on the part of 
the audience. It takes a poor picture indeed to force the audience 
to cheat itself by inattention. 

The program in the theater generally is a well-planned arrangement 
of elements which fit together and which take as long as may reason- 
ably be required to get the desired effect. In broadcasting, because 
of certain administrational problems, the successive elements of the 
evening program are coordinated only with the utmost difficulty, if 
at all, and necessarily run in 15- or 30-minute slices a not always 
convenient or artistic time. At the present time, with the occasional 
obnoxious exception of excessively prolonged or unduly fulsome blurbs 
relative to approaching attractions, the theater screen is practically 
free from advertising, whereas advertising and the sponsored pro- 
gram are at present the commercial basis of the maintenance of broad- 
casting. The elaborate perfection of some feature pictures will be 
duplicable only rarely within the necessary economic limits of broad- 
casting. To the preceding factors may be added the air-conditioning 
of many theaters and the attempts at comfortable theater seating, 
lighting, and the like. All in all, theaters may be expected to be 
attractive places of the public regardless of other entertainment 
media. 

(3) If we consider some deep-seated characteristics of human 
beings, it becomes further evident that the theater has certain ways 
of holding its own alongside a successfully developed television- 
telephone broadcasting set-up. People are interested in change. 
If they are in the home a good deal and most of them are they 
naturally will seek some of their entertainment and diversion else- 
where. The remarkable vogue of the automobile in which many per- 
sons wander rather aimlessly from one place to another largely for the 
sake of motion is a case in point. People are also gregarious and 



July, 1935] TELEVISION AND MOTION PICTURES 45 

somehow seem to have their emotional responses enhanced by crowd 
enthusiasm. One can readily observe this at sporting events, politi- 
cal rallies, revivalist meetings, and other occasions where collective 
enthusiasm or emotional responses are developed. Then too, people 
are distinctly conservative in their pleasures and not prone to aban- 
don hastily anything which for a number of centuries has proved a 
trusted source of entertainment and amusement. It seems most 
likely that the theater and television-telephone broadcasting will 
each be successful fields in their own domain, and that the theater 
need not be unduly apprehensive over the advent of television. 

Nevertheless, it must in all candor be emphasized that film pro- 
ducers and theater managers must not be merely content with past 
achievements. To hold their positions of leadership in their chosen 
fields, they must steadily improve and frequently experiment. It 
is necessary that they shall use whatever good ideas or methods may 
spring from television broadcasting, for example. A merely superior 
or indifferent attitude toward new arts or toward improvements in 
their own older art may prove a first-class passport to diminished pub- 
lic acceptance and ultimate oblivion. Of necessity the motion pic- 
ture industry must also fully avail itself of all the skilled advice and 
guidance which it can secure only from the relatively few experts 
who are acquainted with both the theater and broadcasting. Few 
things would be more dangerous to the motion picture industry than 
dependence upon certain of the pathetically absurd misstatements 
which have been widely circulated by some of its members. How- 
ever, given its natural advantages, a forward-looking attitude, real 
initiative, and careful planning, there appears to be little doubt that 
the motion picture theater can hold an enviable position of public 
acceptance and resulting prosperity in the future as in the past. 



THE THEATRICAL POSSIBILITIES OF TELEVISION* 
H. R. LUBCKE** 

Summary. By analyzing the capabilities of television and of the theater, it is 
shown that they are best adapted to serve different needs. Television is indicated as 
an adjunct to radio broadcasting in creating the "theater of the home" The use of 
motion picture film for television presentations is predicted, and mention of work in 
this direction by the Don Lee Broadcasting System is made. 

The theatrical possibilities of television lie in the future. By that 
is meant the use of television equipment in the theater, as we know the 
theater today. The possibility of television's engulfing and destroying 
the motion picture industry is remote. 

If we examine the capabilities of television and of the motion pic- 
ture, we find that they are best adapted to serve different needs. 
Television, through the agency of broadcasting, is adapted to repro- 
duce a few events for many individuals in many small groups; in 
other words, to reach the public in the home. The motion picture 
and the theater are adapted to reproduce a few events to many indi- 
viduals in a few large groups ; that is, to entertain the public in the 
theater. 

Television can exhibit as many different events simultaneously as 
there are broadcast stations equipped for the purpose, which will be 
comparatively few in number. The theater can produce as many 
plays simultaneously as there are theaters, which is the usual practice. 
We shall look to television as a source of news and timely presenta- 
tions, and to the theater for highly artistic productions of the classics. 

Television can present events as they happen, as well as present 
recorded on enacted versions thereof; and by enlarging its screen 
from home to theater size, it would be possible to present them also in 
the theater. It has been rumored that such a project is under way in 
England, as a private enterprise, and that transmissions are to be re- 
ceived from Ireland. 



* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Don Lee Broadcasting System, Hollywood, Calif. 



46 



THEATRICAL POSSIBILITIES OF TELEVISION 



47 



The great need for television is, however, to bring sight to the 
American radio, so that it may become a complete medium of expres- 
sion, as was accomplished in motion pictures in bringing sound to the 
silent screen. It is in this field that television will develop. And in 
this field its theatrical possibilities will become known. 

The five years of television activity in which the Don Lee Broad- 
casting System has been engaged, and the 
work done by the Radio Corporation of 
America and others, have been directed 
toward producing television broadcasting 
transmitters and home receivers not in 
the development of theatrical television 
equipment. 

The fruition of this activity will bring 
about the "theater of the home," as we 
might term it. This will not be a dupli- 
cation of the present theater. There will 
be no mass psychology, no gathering to- 
gether of people for the common purpose 
of being entertained. The individual will 
be addressed in his home, and his atten- 
tion will be demanded for shorter periods 
of time for fifteen to thirty minutes in- 
stead of from ninety to one hundred and 
twenty minutes. 

The programs will undoubtedly be 
supplied without charge. The combina- 
tion of radio and television comprises the 
greatest advertising medium of all time; 
and as such it will be exploited, regardless 
of cost and other difficulties. It will be 

exploited wisely and on a high plane by the leaders of the broad- 
casting industry. It will be some time, however, before the smaller 
broadcasters enter the field. 

Theatrically, television provides proper expression for drama, 
comedy, news, and what might be called the ' 'short story. ' ' The short 
period of any one program will develop this new treatment on the 
order of a more complete radio drama. Large spectacles, extended 
outdoor plays, and performances of symphony orchestras are not 
suitable for television broadcasting. Their stories can be told in other 




FIG. 1. Sample frames 
from the television version of 
Hawkins and Watkins, In- 
corporated. 



48 H. R. LUBCKE [J. S. M. P. E. 

ways; and the new technic involved will become that of television. 

We shall expect to see theme pictures as we now hear theme songs. 
We shall be interested in them as we now follow the actions of the 
little girl and her fuzzy dog upon the billboards. We shall see institu- 
tional presentations. We shall be educated as to how products are 
made and how processes are carried out. 

The motion picture industry will record many of these programs 
upon film. Film lends itself peculiarly to the production, transmis- 
sion, and distribution of television programs in ways that -are quite 
apparent. Production can be done piecemeal and the result edited. 
The television projector for transmitting motion picture film has been 
brought to the highest state of perfection of any of the transmission 
devices. Positive prints can be sent country-wide for simultaneous 
release, or on a tour of the country in the interests of economy. 

As an example of what has been done in this field, the story for the 
television version of the Mack Sennett comedy Hawkins and Watkins, 
Incorporated, a two-reel talking comedy with Matt McHugh and 
Forrester Harvey in the principal roles, was shortened from twenty 
minutes to eleven minutes. The object was to present the comedy in 
shortened form, so that it could be given, with suitable announce- 
ments, in the standard fourteen and one-half minute radio period. 
The script was reduced from 80 pages of 91 scenes to 8 pages of 15 
scenes. Few scenes were used, to lower the production cost. 

Particular attention was paid to costuming, make-up, lighting, and 
photographic composition. The effort was in the nature of a test to 
find what filming was particularly desirable ; and many principles and 
practices in this direction were established. 

In such manner will theatrical showmanship become a part of tele- 
vision not to address a single audience in a tangible theater, but to 
address the scattered audience of radio not alone to entertain, but 
incidentally and pleasantly to sell the goods and wares of our country. 
Radio, the motion picture, and the stage will contribute to, and be 
engaged by, this new enterprise. Television will become a new enter- 
tainment industry and, as such, can not be considered as a destroyer 
of existing entertainment enterprises. 

DISCUSSION 

CHAIRMAN RACKETT: How long have you been broadcasting motion pictures 
from your local studio? 

MR. LUBCKE: Our television transmitter, W6XAO, officially went on the 



July, 1935] THEATRICAL POSSIBILITIES OF TELEVISION 49 

air Dec. 23, 1931, and we have been broadcasting television programs since that 
time. On May 21, 1932, we demonstrated the first television image ever re- 
ceived in an airplane. A self-synchronized cathode-ray receiver was used. We 
have transmitted more than 7,000,000 feet of motion picture film, through the 
cooperation of the Paramount and the Pathe organizations, including 
Paramount features, Paramount shorts and Pathe Newsreels, as well as certain 
flash news events. Through the courtesy of Pathe News we broadcast scenes 
of the Los Angeles earthquake soon after it happened; and through Paramount 
News, the Stanford-U. S. C. football game on Armistice Day two years ago, 
three hours and forty -five minutes after it was played. 

MR. GLUNT: Within the territory covered by the station, how many persons 
are equipped to receive such broadcasting? 

MR. LUBCKE: The number is in the hundreds rather than in the thousands 
or millions. We have prepared mimeographed information which is sent upon 
receipt of a stamped self -addressed envelope to anyone hearing our signal or who 
is otherwise interested. This gives technical information and references suffi- 
ciently complete so that a somewhat experienced person can construct a television 
receiver. Through such service receivers have been built from time to time by 
various persons, some residing as far east as Pomona, La Verne, and Montebello ; 
others as far north as Santa Paula, some 50 or 60 miles, air line; some within the 
city; and others in other areas surrounding greater Los Angeles. 



MECHANOGRAPHIC RECORDING FOR MOTION PICTURE 
SOUND-TRACKS* 

J. A. MILLER** 

Summary. The mechano graphic system of recording known as the Miller film is 
described. It makes use of a special film having a coating of clear material upon the 
base, which, in turn, is covered by an extremely thin layer of opaque material, approxi- 
mately 2 microns thick. The cutting tool is a specially prepared sapphire, the edges 
of which make a very oblique angle with the surface of the film, cutting a hill-and-dale 
sound-track. The track is cut "clean," the demarcation between the opaque and 
transparent portions being very sharply defined; the opaque portions have a very 
high density, and the transparency of the clear portion is uniform. 

From time to time there have been presented to the Society papers 
dealing with the methods and problems in connection with recording 
sound for motion picture accompaniment. In the beginning these 
papers had to do with recording on wax and of late with the photo- 
graphic processes of recording on film. There is still a third method 
which has not yet been described before this Society and which is 
destined to play an important role. The introduction of this method 
is the reason for this paper. 

It is perhaps best that we review briefly the history of sound re- 
cording so we may all know the long and tedious paths that have 
led up to the present stage of the art, and thus better visualize what 
is apt to happen in the next few years. The first sound recording 
and reproducing machine was built by Edison in 1876, which makes 
the industry now 60 years old. This first machine was a metal 
cylinder mounted upon a shaft with a flywheel and a hand-crank for 
turning. The surface of the cylinder was grooved, and over it was 
stretched a thin sheet of tin-foil to receive the recording. The 
sound box was of the hill-and-dale type and the needle did the re- 
cording at the same time that the tin-foil was pressed into the groove 
of the cylinder. Work upon the machine was discontinued from time 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** American Mechanographic Corp., New York, N. Y. 

50 



MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 



51 



to time on account of the development of the electric lamp, and it 
was not until the 80's that Edison perfected the well-known Edison 
phonograph. 

It is interesting to note that at this time C. E. Fritts was working 
upon a method of recording sound upon film photographically, and 
in 1880 he filed his first patents. These patents were issued in 1917, 
long after the inventor's death and after being in the patent office for 
the record time of thirty-seven years. 

In 1891 the mechanographic system was invented, and in 1898 
came the telegraphone, or magnetic system. 

Thus it will be seen that from the beginning of the present century 



Hill & Da/e Record J3% RPM 
600^ f 600Ocps. 
tfeecf/e point ro cfius. .oo/s " 
/ns/de groove 6 ~D. 




^fy/5 represents a defect oooy cf/o. 



0-^t IQOOOcps. 



^^L^ 

FIG. 1. Showing deformation of surface of record by high unit-pressure at 
point of contact, where rapid changes occur in the pressure angle, and de- 
fects causing surface noise. 

we have been engaged only in perfecting the processes that were in- 
vented much earlier. Although patents have been granted on other 
systems they have never reached a commercial stage, so they will be 
ignored. 

The four present-day commercial methods may be defined as 
follows : 

Mechanical Wherein the recording and reproducing are done mechanically. 

Photographic Wherein the recording is done photographically and the re- 
producing is done optically. 

Mechanographic Wherein the recording is done mechanically and the reproduc- 
ing is done optically. 

Magnetic Wherein the recording and reproducing are both done magnetically. 



52 



J. A. MILLER 



[J. S. M. P. E. 



The progress toward perfection of these methods can be traced 
with respect to the perfection of materials and the reduction of physi- 
cal limitations of processes and equipment. It must be borne in 
mind that in recording sound a measurement of one ten-thousandth 
of an inch is a large measure, and other physical characteristics must 
be correspondingly limited. Many times it has been necessary to 
await the invention of auxiliary equipment, such as the vacuum-tube 
amplifier. 

With primitive equipment it was much easier to arrive at, say, a 
90 per cent result by the mechanical method than by any of the others. 
That is the reason why it led the field in development and commercial 



Hill*Da/e Record/ 20 yrpnncft 



Lateral ftecord 66 yr:f>/: 




/oooo c/>s. Ji ftPM 6"/a.oroove 7000 c/>s. 

FIG. 2. Illustrating shock excitation and cause of distortion. 

adaptation. With present-day equipment, however, it is easy to 
attain a 90 per cent result with any of the four systems listed above; 
but as far as motion picture recording is concerned, the mechanical 
(or disk) system has been eliminated and the telegraphone is not 
adaptable, so we can consider principally the other two systems, the 
photographic and mechanographic. Under the heading of photo- 
graphic we must include the variable-density and the variable- width 
systems of recording, in considering the limitations of each. All 
the systems have inherent limitations, and perfection becomes in- 
creasingly difficult with each additional improvement; and further- 
more, consideration must be given to the adaptability of the system 
to the purpose at hand. 

In magnetic recording a hard steel wire is used as the record ma- 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 53 



terial. As the wire is run through a steady 
magnetic field, the recording magnets act to 
fix the molecules in a definite state. The fixed 
magnetic field can be compared to the C-bias of 
a vacuum-tube. In order to produce good 
quality it is necessary that the recording mag- 
nets be in contact with the wire. The limi- 
tations of the system are, therefore, the mag- 
netic hysteresis of the wire and the mechanical 
irregularities of the surface of the wire, which 
produce ground-noise. There is also the diffi- 
culty that a high modulation at a given stretch 
of the wire will be transferred to an adjoining 
turn of wire, thus appearing as an echo. The 
quality of the recording can be quite good, but 
in order to keep the surface-noise down the 
preparation of the wire becomes very expensive. 
The wire is very difficult to repair when it 
breaks and, in general, the system is adaptable 
to very few kinds of work. 

With mechanical recording we have witnessed 
a most peculiar development, wherein Edison 
very early in his work arrived at a hill-and- 
dale recording upon wax, with an acetate press- 
ing; and now after fifty years of intense research 
we have arrived back exactly at the same point. 
Adding to the descriptions in many of the early 
Edison patents a chapter on modern vacuum- 
tube amplifiers, one has the whole story. The 
limitation of the mechanical system lies mainly 
in the playback mechanism. The recording is 
done with a sapphire having a straight-line cut- 
ting edge, whereas the reproducing is done with 
a ball-pointed needle. If a small pointed needle 
is used, the radius of the ball may be one and 
one-half thousandths of an inch. Of course, if 
a hard record is used with a steel needle, the 
point endures only for about two revolutions; 
but if a sapphire or diamond is used upon acetate, 
then the point remains, and the surface of the 





(b) 





(C) 



FIG. 3. Illustrat- 
ing the limitations 
of photographic re- 
cording: (a) 6000 
cps., full modula- 
tion; (b) 6000 cps., 
modulation reduced 
25 db. from sound 
level of a; (c) un- 
modulated. 40x. 



54 J. A. MILLER [J. S. M. p. E. 

record is deformed by the extremely high unit pressure at the 
point of contact, especially at points where rapid changes occur 
in the pressure angle as shown in Fig. 1. The change of the 
pressure angle causes rectification of the main wave and shock- 
excitation of secondary degrees of freedom of the moving element 
of the reproducer, giving a very sharp and unnatural effect to 
the high frequencies. Shock excitation may occur also in the 
record material itself: the elasticity of the material can not be 
ignored at the higher frequencies. In addition, the surface of the 
material is not completely homogeneous, and thus causes surface- 
noise (Fig. 2). 

With variable-density film recording, the limiting factors are 
numerous, such as lens dispersion, width of recording slit, fogging 
and spreading of the image in the emulsion. In addition, there are 
critical factors that must be accurately met, but can not be listed di- 
rectly as limiting factors, such as light exposure, uniformity of the 
emulsion, matching emulsions from negative to positive, printing 
exposure, development time, etc. 

With variable-width recordings the limitations are lens dispersion, 
film image spreading, fogging, emulsion irregularities and width of 
the recording slit. In fact, with these two systems Kellogg's 1 
simile can not be improved upon: "We have given our painter not 
only too big a brush but a piece of blotting paper upon which to 
make the picture, and paints that run" (Fig. 3). 

The physical limitations of the photographic processes have been 
so reduced during the past five years that it is really remarkable what 
can be achieved by these methods, which were regarded as hopeless 
as late as 1928; but it is unfortunate that such a large proportion of 
recording is so far below the top level that has been attained. The 
reasons for poor and medium quality recordings are not easily as- 
scribed to their proper source, and consequently there has grown up 
the familiar chorus we have all heard so often: the placement was 
wrong, the piano was bad, the recorder was wrong, the film was bad, 
the laboratory spoiled it, the gamma was wrong or perhaps the repro- 
ducer was "haywire." One of these things doubtless was at fault; 
but it is a long road to travel before one finds out he is on the wrong 
road. 

With these things in mind we survey the field to see exactly what 
it is we want in the way of a recorder for a motion picture track, and 
then check against what is available. First, as has been proved by 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 55 

practice, the sound-track must be upon the same film with the pic- 
ture. This eliminates from consideration both magnetic and me- 
chanical systems. Now, assuming the same quality among the other 
methods, I should say choose a variable-width track, because the 
critical conditions of recording are more easily fulfilled, and it is much 
safer to transfer photographically from the negative to the positive. 
This is especially important where dupe negatives are used in the 
foreign markets. The next thing that is necessary is an immediate 
playback, and by this I do not mean the playing back of some make- 
shift record taken on an entirely different recording system, but the 
immediate playback of the finished record which is to be used for 
production purposes. The full importance of this is not realized 
until it is seen in actual practice. So far, the director of a picture 
has not been the director of his sound ; and in many instances he has 
had to make the best of a bad situation, which could have been 
avoided by checking the recording directly after it was made. Not 
only can the quality of the picture be improved by due application 
of this principle, but the cost of production can be reduced also. In 
order to make this point concrete, mention might be made of two 
pictures which were released with an interval of time between, in 
which the same operatic star appeared, the first picture a failure 
and the second a grand success. Non-technical persons whom I 
have interviewed have agreed that the sound of the first was bad and 
that of the second exceptionally good, whereas technical men have 
told me that the first was good and the second so bad that changes in 
the reproducing equipment were necessary before the picture could 
be shown in the theater. This is only one example where sound 
may have been the cause of failure of a picture. There are other 
more outstanding examples and a close study of the situation might 
be very surprising. In view of this fact the author wishes to state 
that in his opinion it is the duty of the engineers to place into the 
hands of the director of the picture the tools best suited to his trade, 
and not to assume the responsibility for the psychological coordina- 
tion of sound and picture: at the earliest possible moment the en- 
gineer should make it possible for the director to assume this responsi- 
bility. The director is endowed with a temperament more closely 
attuned to that of the public than is possible of attainment by a 
technical person, and it is therefore better business for everyone con- 
cerned to rely upon the director in this respect. It should not be 
inferred from this that the importance of the recording business as 



56 J. A. MILLER [j. S. M. p. E. 

such should be under-rated; in fact, recording should not be regarded 
simply as a "spare wheel" for the motion picture industry but as an 
essential and cooperating art, older in years, but as yet drawing only 
two cents of each motion picture dollar. 

The next step in our survey then brings us to the mechanographic 
system as the only means of achieving all the desired results de- 
scribed. By means of this method it is possible not only to play 
back the completed record as soon as it is recorded, but to play the 
record while it is being recorded, with a delay corresponding to four 
frames of the picture. We are also able by this method to reduce the 
paint brush to which Kellogg referred to an irreducible minimum and 
to eliminate the blotting paper. 

The first mechanographic machine was built in Berlin in 1891, and 
consisted of a paper strip coated with a black layer into which was 
cut a lateral groove, thus removing the black layer within the groove. 
Upon this groove was cast a strong light, which passed through the 
paper and through a lens which focused a half -image of the track 
upon a slit, behind which was placed a selenium cell which in turn 
operated a telephone receiver. Considering the materials available 
at that time it will be realized that that was a big step: but with 
present-day knowledge it is easy to see how the physical limitations 
of the equipment and materials would admit of only the most primi- 
tive results. Nothing further seems to have been done with the sys- 
tem until 1930, when Berthon reinvented it, and in conjunction with 
Nublat, developed a machine that was put upon the European mar- 
ket as a phonograph, and known as the Nublat machine. The later 
models of this machine, instead of cutting into a black layer, were 
made to cut through the center of a thin black film, thus making a 
sound record upon the edge of each half when the film was split. 
Printing was done with cylindrical optical enlargement in order to 
increase the amplitude of modulation. The results attained were 
only fair, as might be expected on account of the severe physical 
limitations. It will be seen that at a frequency of 6000 cycles and 
at standard motion picture speed, and with a cutting tool having a 
heel angle of 45 degrees, it is possible to cut to an amplitude of only 
0.002 inch, whereas an amplitude of 0.080 inch is required for mo- 
tion picture film. At the same time, if it were possible to reach 
practical amplitudes the power required would be enormous; in 
fact, several kilowatts would be required to record up to a frequency 
of 10,000 cycles per second. Although the machine was introduced 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 57 

upon the phonograph market in France, the quality was not good 
enough for the motion picture field. 

It was not until 1931 that the invention was made which now makes 
it possible to attain with the mechanographic system better results 
than with any other system. The system now used, known as Miller - 
film, makes use of a special film having a coating of clear material 
upon the base, which, in turn, is covered by an extremely thin layer 
of opaque material, approximately two microns thick. For all or- 
dinary purposes the film is very tough and durable : in fact it can not 
be scratched by the fingernail. The cutting tool (Fig. 4) is a specially 




- -O80" 

cutter 




FIG. 4. The cutting tool used in the mechanographic system. 

prepared sapphire, the edges of which make a very oblique angle 
with the surface of the film. Instead of cutting a lateral track as 
heretofore, the track is hill-and-dale. It will now be observed 
(Fig. 4) that a very small movement of the cutting tool in the verti- 
cal direction produces a great change in the quantity of black layer 
that is removed. Thus is attained a mechanical amplification of 
from fifty to one hundred times, comparing the movement of the 
cutting tool to the variation of the width of track cut. Instead of 
having to move the cutting tool 0.080 inch for full modulation of 
the sound-track, it is now possible to attain full modulation with a 
tool movement of 0.001 or 0.002 inch. This forms the basis of the 



58 



J. A. MILLER 



[J. S. M. P. E. 



(a) 



FIG. 5. Compari- 
son of lines of de- 
marcation between 
opaque and trans- 
parent sections of 
record ; (a) photo- 
graphic negative; 
noise reduction, zero 
modulation, variable- 
width; (6) mechano- 
graphic; zero modu- 
lation ; (c) photo- 
graphic positive; low 
modulation. 300x. 



development, but in order to bring it to the 
point where it could be applied commercially 
it was necessary to cover much unexplored 
territory. In the first place there was no 
film that could be used, even for a test; 
no cutters were available that would answer 
the requirements; and no machine that would 
carry the load necessary for this work. In 
fact, the only thing available in sufficient 
quantity was advice to the effect that the 
system could never be made to work, and 
that was present in abundance. 

Many tests were conducted to determine 
what material could be applied to a film that 
would be flexible, transparent, and have the 
same cutting characteristic as wax. Upon the 
surface of such a layer must be placed an 
extremely thin layer of opaque material hav- 
ing a fine grain structure which will cut with 
a smooth surface, have sufficient strength to 
withstand damage, and have a definite line 
of demarcation between the opaque and trans- 
parent sections. This was a complete research 
problem in itself, but film is now available that 
fulfills all these requirements; and when it is 
cut, as can be seen in Fig. 5, the line of 
demarcation is much more definite than can 
be achieved photographically. The importance 
of this fact must not be overlooked, as the 
limitation of all systems depends upon the 
definiteness of this line of demarcation on an 
unmodulated track. If the maximum modu- 
lation of the track is of 0.080 inch, and the 
desired range 125 db., then the first 100 db. 
must be included within modulation peaks not 
exceeding 0.004 inch high, which does not 
allow for much irregularity in the line of de- 
marcation. It is possible to obtain film of this 
sort with the black pigment in colloidal form 
and of such concentration that grain size is no 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND -TRACKS 59 

longer a factor as in the photographic process. The edge to which 
a sapphire cutter can be ground is microscopic; whence there is no 
limitation that corresponds to the width of the recording slit, so that 
frequencies as high as 25,000 cps. can be recorded (Fig. 6). Of course, 
in reproduction, the width of the reproducing beam enters as a detri- 
mental factor, but in this case it is not so important as in recording, 
and is far better than any needle could be. Irregularities of the 
emulsion are no longer present, and even the surface of the film 





(a) 



FIG. 6. 6000-cycle recordings; original negatives: (a) 
sound-track; (6) mechanographic sound-track. 



photographic 
100 x. 



where it has been cut is much improved over a plain emulsion 
surface. 

The cutter was the next item of importance, and required a wide 
departure from current practice inasmuch as it has to be a constant- 
amplitude device. In order to gain sufficient movement at high fre- 
quencies it is necessary to tune the element to a high audio frequency. 
A cutter has been produced that will record up to 10,000 cycles per 
second with a power consumption of about two watts. From this it 
is easy to understand the impracticability of the old mechanographic 
system with a lateral groove. The movement of the element would 
be fifty times greater, whence the requisite power would be in pro- 



60 



J. A. MILLER 



[J. S. M. P. E. 




FIG. 7. Illustrating the advantage 
gained in noise-reduction due to irregu- 
larities in the transparent portion of the 
film, and the sharp demarcation in the 
mechanographic recording: (a) SMPE 
standard, tips, 6000 cps.; (6) SMPE 
standard, base, 6000 cps.; (c) mechano- 
graphic sound-track at point where 7000 
cps. occurs. 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 61 



portion to the square of this figure, or 
2500. In addition, the efficiency would 
be only one-tenth as great, on account of 
the additional air-gap, which would make 
the power ratio 25,000 to one. In other 
words, instead of two watts, the cutter 
would require about fifty kilowatts. The 
details of construction of the cutter can 
not be given at the present time, but the 
response curve of a typical example can 
be constructed flat within two db. for 30 to 
8000 cycles or from 30 to 10,000 as 
desired. 

The machine itself offered several prob- 
lems inasmuch as the work done upon the 
film by the cutter must not interfere with 
the uniformity of movement of the film. 
An ordinary sized flywheel prevents any 
variation in speed of the driving sprocket, 
but special provision must be made to 
prevent slippage of the film on the 
sprocket. Several means of doing this 
have been developed. 

A study of an original record of the 
process leads to many interesting observa- 
tions. More than one track can be 
recorded upon the same piece of film. 
Cutting and re-recording can be done 
from the original. Short ends can be 
utilized. There is no darkroom loading 
of magazines. Examination of the track 
shows that the groove is cut clean, the 
surface is as smooth as glass, the opaque 
portions have a density too high to mea- 
sure, and that the transparency of the 
clear portion is uniform. 

Experiments with film of varying de- 
grees of resolution leads to the conclu- 
sion that, whereas the fundamental tone, 
the power, and, to some extent, the 




FIG. 8. Photographic 
copies, or prints, (a) from 
photographic sound-track ; 
(b) from mechanographic 
sound-track ; (c) from photo- 
graphic sound-track. 300x. 



62 J. A. MILLER [J. S. M. P. E. 

character of speech are governed by modulations of high amplitude, 
the realness or lifelike quality is governed by modulations of 
very low amplitude. That is, a voice can be recorded loud 
and clear, with perfect intelligibility, but it may be still unrecogniz- 
able as to the original speaker. This is also seen in a recording in 
which there is room tone or echo. If the amplitude of the reverbera- 
tions is high, they cause confusion, but if their amplitude is low and 
they lie in the valley between high amplitudes of the original sound, 
they add character to the sound. In the case of a spoken word the 
time occupied by these low amplitude sounds may be fifty per cent of 
the total time taken for the word. This time-interval varies through 
a wide range for various voices, so that a voice with a high ratio of 
large amplitude sounds is an easy voice to record, while the converse 
is also true if the low modulations are lost on account of the poor 
resolving factor of the film. This experiment has been performed 
with film having an opaque layer of varying degrees of uniformity, 
and it is found that the naturalness decreases steadily as the line of 
demarcation becomes less and less definite. The same corresponding 
result is arrived at if a print is made from a good negative onto or- 
dinary photographic positive film. This is the factor that deter- 
mines the minimum recording level, and is evident when a recorded 
sound is so disturbed by ground-noise that it is annoying for the ear 
to separate the two. The useful recording range of the system is 
then the ratio between this minimum sound level and the maximum 
sound level that the track can accommodate. The perfect recording 
system would encompass about 125 db., but with the present photo- 
graphic systems the range varies between 35 and 45 db. and with 
hill-and-dale acetate records from 50 to 55 db. Therefore, in making 
measurements of losses at high frequencies, either from recording or 
printing, it is not sufficient to say that the loss is six or eight db. on a 
fully modulated track; the recording range and the loss in the last 
ten db. of the recording range toward the end of minimum recording 
level must be stated also. 

Noise reduction methods reduce the disturbance introduced by 
irregularities occurring in the transparent part of the track on a dirty 
film, but have no effect upon limitations caused by irregularities of 
the materials of the record carrier at the line of demarcation. Eight 
or ten db. are gained in this manner if the film is bad, but any real ex- 
tension of the range must be accomplished by improving the material 
of the carrier itself. That this improvement is accomplished by the 



July, 1935] MECHANOGRAPHIC RECORDING FOR SOUND-TRACKS 63 

mechanographic system is clearly seen in Fig. 7, and it should be pos- 
sible to have a consistent working range approaching 75 db. by this 
method. This can be appreciated better by realizing that if the same 
degree of perfection of the record carrier is arrived at as is reached 
with an acetate record of the hill-and-dale type, then the advantage 
gained is of the order of 50 db., which would indicate a possible range 
of 100 to 105 db. We assume first that all major imperfections are 
easily removable and that the ones most difficult to remove are of an 
order of magnitude of 0.0005 inch or less. In a hill-and-dale me- 
chanical record this irregularity produces 100 per cent modulation at 




FIG. 9. Mechanographic sound-track showing reso- 
lution at 50,000 cps., at a recording level 55 db. below 
maximum. 

6000 cycles, whereas on the mechanographic record it would produce 
a disturbance 50 db. below full modulation on account of the differ- 
ence between constant-velocity and constant-amplitude recording. 

In other words, when the time arrives for sound to be brought up 
to the standard that is now being demanded by a large part of the 
public, it will be necessary to re-record mechanographic ally on each 
release print. Of course, in the meantime photographic copies of 
mechanographic track can be used, thus obtaining a result that lies 
between the present photographic method and the direct recorded 
mechanographic (Fig. 8). 

In conclusion, in Fig. 9 is seen a sample of resolution at 50,000 
cps., 55 db., below maximum, that is better than can be achieved 
by photographic methods at 5000 cps. 



64 J. A. MILLER 

REFERENCE 

KELLOGG, E. W.: "The Development of 16-Mm. Sound Motion Pictures," 
/. Soc. Mot. Pict. Eng., XXIII (Jan., 1935), No. 1, p. 65. 

DISCUSSION 

DR. FRAYNE: Mr. Miller stated that the director should be the director of 
the sound as well as of the picture. At the present time I believe he does direct 
both. But to carry Mr. Miller's proposed schedule to its conclusion, the director 
should be able to view the picture immediately also. I hesitate to make any 
suggestions as to how that might be done. 

PRESIDENT TASKER: How did you make the 50,000-cycle recording? 

MR. MILLER: The recording was not done at 50,000 cycles; it was done 
at 8000, letting the machine slow down. It was done simply to show the resolu- 
tion on the film at 50,000 cycles. Below that frequency, good wave-form is 
still maintained, which is very essential. 



THE KODACHROME PROCESS FOR AMATEUR CINEMA- 
TOGRAPHY IN NATURAL COLORS* 

L. D. MANNES AND L. GODOWSKY, JR.** 

Summary. The new Kodachrome reversal process is a three-color subtr active 
process, the colors being formed by combining the three complementary or "minus" 
colors, which absorb the corresponding primary colors from the projection beam. 
There are three emulsion layers, separated by thin layers of clear gelatin, the first 
layer being sensitized to blue, the second to green, and the third to red light. 

The film is exposed in the ordinary manner, with the emulsion toward the lens. 
Processing is done by the reversal process, and the three positive images are their differ- 
entially dyed the three corresponding minus colors. The finished Kodachrome film 
is projected in the usual manner, without filters. 

INTRODUCTION 

The Kodachrome reversal process recently introduced is the result 
of an attempt to produce a color process that would involve no prob- 
lems not incidental to black-and-white photography. While this was 
by no means achieved in its entirety, the process is, at least from the 
photographers' and projectionists' points of view, as simple as black- 
and-white photography. The problems involved are confined to the 
manufacturing and processing of film, thus placing the burden upon 
highly organized production facilities rather than upon the sometimes 
unskilled consumer. Pictures taken by this process are exposed at 
virtually normal speed in an ordinary 16-mm. camera of any type 
having a capacity of 100 feet and projected with any 16-mm. projec- 
tor. No filters are usually necessary in taking or projecting, save in 
the case of artificial light, where a blue compensating filter is used 
over the camera lens. Under special daylight conditions involving 
haze, cold light, or an abundance of reflected ultraviolet light, it is 
sometimes desirable to expose through a colorless filter which absorbs 
most of the ultraviolet but does not change the color balance or 
exposure conditions. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. Communica- 
tion No. 549 from the Kodak Research Laboratories. 
** Eastman Kodak Co., Rochester, N. Y. 

65 



66 L. D. MANNES AND L. GODOWSKY, JR. [J. s. M. P. E. 

I. CLASSIFICATION OF THE KODACHROME PROCESS 

Color processes are generally divided into two types, the additive 
and subtractive. With the additive process the actual red, green, 
and blue colors are either visible in the film itself or are formed by an 
optical system, as with Kodacolor. These primary colors are pro- 
jected onto the screen for viewing. Additive methods include Koda- 
color, color-screen plates (such as the Autochrome plate), Duf ay- 
color, and others. 

Any possible color upon the screen can be formed by additive com- 
binations of the primary colors, red, green, and blue. Likewise, they 
can be formed by proper subtractive combinations of the comple- 
mentary colors. The complement of red is blue-green, of green is 
magenta, and of blue is yellow. These complements are sometimes 
referred to as the minus colors, thus : 

Primary Color Complementary Color 

Red Blue-green (Minus red) 

Green Magenta (Minus green) 

Blue Yellow (Minus blue) 

Subtractive processes, of which Kodachrome and Technicolor are 
examples, form their colors by combining the three complementary 
or minus colors. The minus or subtractive colors merely absorb the 
corresponding primary color from the projection beam of light. Such 
a light-beam may be considered as white light or as containing a com- 
bination of all colors. 

If we put a blue-green (which is to say, minus red) filter or dye de- 
posit into such a white light-beam, the red will be absorbed and the 
screen will appear blue-green. If, however, we add a magenta (minus 
green) filter to the blue-green filter mentioned above, both the red 
and the green light will be subtracted from the white light-beam. 
The only light left to go on through to the screen is blue, and that will 
be the color of the screen. Similarly, any color may be formed. 

Accordingly, the blue-sensitive layer of Kodachrome will in the 
finished positive contain the complement of blue, which is yellow; 
the second or green-sensitive layer, the complement of green, namely, 
magenta; and the bottom or red-sensitive portion will contain the 
complementary color of red, which is blue-green. 



July, 1935] KODACHROME PROCESS IN NATURAL COLORS 67 

II. FILM 

Kodachrome film is panchromatic and bears the ordinary jet back- 
ings as used on regular Cine*-Kodak film. The emulsion, however, 
consists of three layers, each sensitized to one of the primary colors 
and separated from the adjacent layer by a thin coating of clear gela- 
tin. The top layer of emulsion, upon which the light first falls in ex- 
posure, is sensitive only to blue light, but it does transmit green and 
red light to the layers underneath. While it is sensitive to the blue, 
it also contains a yellow dye which prevents the blue light from pass- 
ing through to the silver bromide grains below. 

The second or middle layer is sensitive to green and blue light, but 
as all blue is filtered out by the yellow dye just mentioned, we need to 
consider only its reaction to the green. 

Next to the film support is the bottom or third emulsion, which is 
sensitive to red and blue; but here again, the blue being stopped in 
the surface layer, this emulsion reacts to red only. 

Briefly, there are three separate emulsions sensitized as follows: 
the first or top layer to blue, second to green, and the third or bottom 
to red light. Each coating is exceedingly thin so that the total thick- 
ness is about the same as the thickness of black-and-white coatings. 

m. EXPOSURE 

Kodachrome film is exposed in the normal manner, that is, with 
the emulsion side toward the lens. It is unnecessary for the light to 
pass through the support, as it does in Kodacolor or any of the color- 
screen processes. The speed of the film is somewhat less than that of 
regular panchromatic Cine-Kodak films. 

IV. PROCESSING 

Processing is carried out in continuous machines by a reversal 
process which converts, by the usual method, the images in all three 
layers to their corresponding positives. These three positive images 
are then differentially dyed the three corresponding or minus colors 
previously described, i. e., blue-green, magenta, and yellow. 

All silver is then removed from the film, after which it is washed 
and dried. The final positive accordingly carries a dye image only. 

From the three stages of processing described, we now have the 
three complementary colors in their respective layers. The amount 
of each will depend, of course, upon the amount of silver bromide 



68 L. D. MANNES AND L. GODOWSKY, JR. 

removed in bleaching the original negative. The three colors in turn 
will combine to form the positive image in natural color. 

v. PROJECTION 

As the finished Kodachrome film contains the actual colored image, 
projection is carried out under normal conditions. No filter is used. 
The light-source, projection distance, and type of screen are the same 
as for black-and-white film. 

DISCUSSION 

MR. KELLOGG : Will Kodachrome be available in roll film for ordinary cameras? 

MR. CRABTREE: I do not know; it is not available at the present time. 

MR. DUB RAY: What would be the effect of under- or over-exposure? 

MR. CRABTREE: The effect is relatively the same as with black-and-white 
16-mm. film. 

MR. HOPPER: How permanent are the colors? Is the dye fairly permanent 
after projection? 

MR. CRABTREE: I see no reason why they should not be satisfactorily per- 
manent. Any dye, of course, is apt to fade under certain adverse conditions. 

Miss EVANS: How soon will you be able to develop prints on the Pacific 
Coast without sending the films back to Rochester? 

MR. HUSE: It is contemplated that by late fall we shall have equipment 
available in Hollywood for processing Kodachrome film, just as we have for black- 
and-white films. Arrangements are being made at the present moment to that 
end. 

Some persons may have been wondering what is going to happen as regards the 
35-mm. film. Up to the present time all our developmental work has had to do 
with the 16-mm. process. A great deal of work is still to be done in making 
duplicates or prints from that. Our experimental work on 35-mm.' film has not 
really begun. We shall, no doubt, collaborate with Technicolor on that, for the 
reason that Dr. Troland's patent indicated that he had done work on a somewhat 
similar process. What will happen in the future we must wait to find out. 

MR. TIMMER: Is it possible to put a sound-track satisfactorily on this film? 

DR. SANDVIK: There is one problem that should be considered, and that is 
the problem of getting dyes that will, in addition to giving satisfactory visual 
color rendering, also have sufficient absorption in the spectral range to which 
the photoelectric cell is sensitive. 

As most of you know, in the past two years the trend has been to extend the 
sensitivity of photoelectric cells farther and farther into the infrared, until now 
some of the cells have the maximum sensitivity at 0.9 mu, which is far beyond 
the red end of the spectrum. It is a question then of either choosing dyes having 
a fairly high absorption in the infrared or else sacrifice some of the photoelectric 
cell sensitivity by using dyes that have a lower absorption throughout the spectral 
region to which the photoelectric cell is sensitive. 



INTRODUCTION TO THE PHOTOGRAPHIC POSSIBILITIES 
OF POLARIZED LIGHT* 



F. TUTTLE AND J. W. McFARLANE** 



Summary. The introduction of an efficient plane polarizing sheet material in 
sizes large enough to cover lenses and lights has made simple the use of polarized light 
in photography. An Eastman Pola-screen, incorporating this material, over the 
lens, allows unusual sky effects, photographing obliquely through glass and water 
without reflections, and photographing other surfaces obliquely to show surface detail. 
When the subject is illuminated through larger Pola-screens, in addition, complete 
control of gloss results. Faces so photographed can appear unnaturally perspiry, or 
devoid of all lustre, depending upon the camera Pola-screen position. Reflections 
from animation cells can be greatly reduced, and photographing any small subject 
that presents a reflection problem is quite simple. Various trick lightings and color 
effects are also attainable. 

Our eyes respond naturally to differences in color and in intensity 
of light, and it is by these differences that we are able to see the world 
around us. There is another property in which light rays may differ, 
but our eyes, unaided, can not see those differences. This property 
is called "polarization," and is concerned, as explained later, with the 
manner in which the light ray vibrates. Light rays may be polarized 
by optical devices; they are also partly polarized by reflection from 
common objects. It also happens that clear skylight is partly po- 
larized. For the last two reasons, much of the light by which we see 
things is polarized to some extent, a fact we first realize when we 
look through an Eastman Pola-screen. 

HOW LIGHT WAVES ARE POLARIZED 

While the nature of light is not entirely understood, many phe- 
nomena can be explained by assuming that light is a vibratory motion 
which is propagated through space in the form of electromagnetic 
waves. Among these phenomena is the polarization of light. 

The vibration of a light wave is not along the direction of the ray, 
as in the case of sound, but is at right angles to the ray and usually 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Eastman Kodak Co., Rochester, N. Y. 



70 F. TUTTLE AND J. W. McFARLANE [J. S. M. P. E. 

. 

in all possible directions, that is, up and down, sidewise, etc. (Fig. 1). 
It is possible by various devices to change the light ray so that it vi- 
brates in only one direction, as shown in Fig. 2. This one vibration is 
not only composed of the one originally vibrating in this same direc- 
tion, but is also composed of parts of all the others, except the one vi- 
brating at right angles to it. The result is that almost half the light is 
allowed through, even though there is only one direction of vibration. 
It may help to explain this if we draw an analogy : 

Imagine a string stretched horizontally, passing through a slit in 
a card at right angles to it. If the slit is vertical, the string is able to 




SOUND VIBRATIONS 




LIGHT VIBRATIONS 

FIG . 1 . Sound waves vibrate along the 
direction of travel; light waves vibrate 
at right angles to the ray and, ordinarily, 
in all possible directions. 

vibrate in a vertical plane only, and if the slit is horizontal, the vi- 
brations are restricted to a horizontal plane. If the card is rotated, 
the vibration plane of the string follows the slit. Light behaves 
in much the same way, except that the vibrations require the optical 
equivalent of a mechanical slit. A light ray in which only one di- 
rection of vibration exists is said to be plane polarized, that is to say, 
polarized in one plane. The vibration plane, that is, the plane 
parallel to the vibration of the emerging ray, is definitely fixed in the 
polarizing device, and is rotated when the polarizing device is ro- 
tated. A second polarizing device, placed in the path of the ray 
leaving the first polarizer, may or may not transmit the plane polar- 
ized ray, depending upon its angular position. At one angle, prac- 



July, 1935] PHOTOGRAPHIC POSSIBILITIES OF POLARIZED LIGHT 71 

tically no light is allowed through, and the polarizers are said to be 
"crossed." At 90 degrees from this position, all the light from the 
first polarizer goes through, and the polarizers are said to be 
"parallel." 

A number of polarizing devices, such as prisms made from crystal- 
line Iceland spar, known as Nicol prisms, have been known to sci- 
entific workers for many years. Up to now, there has been no 
polarizing device suitable for ordinary photography. Nicol prisms 
are very costly, have a very small field, their length being much 
greater than their free aperture. The desirability of a highly efficient 

PLANE. 
POLARISED RAV 




UNPOLAR1ZE.D 
RAY 



PLANE. POLARIZING 
DEVICE, 



FIG. 2. Any plane polarizing device cuts down the 
possible directions of vibration to only one; this one is 
parallel to the vibration plane of the device. 

polarizing device in sheet form has been recognized for some time. 
Such a material is now available, having the necessary optical prop- 
erties and capable of being produced upon a commercial basis in 
sufficiently large sheets at a cost low enough to make it practical in 
photography. The invention is due to Mr. Edwin H. Land, who 
was the first to prepare a commercially practical sheet containing a 
polarizing material oriented properly for satisfactory performance. 
(U. S. Patents 1,918,848, 1,951,664, 1,956,867, 1,989,371.) In this 
material are countless minute rod-like crystals, which are all parallel 
to each other. They may be regarded as optical slits, so when the 
material is rotated, the direction of light vibration is rotated, just 
as rotating the slit made the string's direction of vibration follow 
it. The Eastman Pola-screen incorporates this polarizing sheet 



72 



F. TUTTLE AND J. W. McFARLANE [J. S. M. P. E. 



material, cemented between glass plates. The Eastman Pola-screen 
is therefore a large polarizing device, free from the limitations of the 
Nicol prism. 

THE IMPORTANCE OF POLARIZED LIGHT IN PHOTOGRAPHY 

The importance of polarized light in photography is due to the way 
in which all natural substances reflect polarized light. When a ray 



SUBJECT- 




DIFFUSELY REFLECTED 
LIGHT SPECULARLY " 



CAMELRA. 



FIG. 3. Photography by diffusely reflected light, using polar- 
izing equipment. Light reflected specularly retains its polarized 
form; it may therefore be cut out by a Pola-screen at the camera, 
/ indicates a Pola-screen, Type I ; // indicates a large Pola-screen, 
Type II. The indexes on the two Pola-screens show their planes 
of vibration. 

of light falls upon, for instance, a sheet of paper, the light that is re- 
flected is composed of two parts which are technically known as the 
specular and diffuse components. The specular component produces 
what we know as gloss and enables us to see more or less distinctly an 
image of the source of light. Light reflected from polished metallic 
articles is almost entirely specular, whereas that reflected from chalk 
is almost entirely diffuse. The diffuse component is reflected in all 
directions, and hence does not give rise to an appearance of glossiness. 
Now, if the ray of light that is illuminating our subject is plane 



July, 1935] PHOTOGRAPHIC POSSIBILITIES OF POLARIZED LIGHT 73 

polarized, the reflected rays that form the specular component are 
still plane polarized, but the rays reflected diffusely are not. If we 
look at the subject through a Pola-screen, we can orient the screen 
so that practically all the specular reflection is stopped, and see the 
subject by diffusely reflected light. This fact, which is extremely im- 
portant, permits the many applications described below. The use 
of Pola-screens in front of the lights illuminates the subject with 
plane polarized light. Another such device at the camera lens per- 
mits photographing by the diffusely reflected light alone, as is shown 
in Fig. 3. This is desirable in many cases, because the specularly re- 
flected light or glare obscures more or less the detail that it is desired 
to record. It is obvious from Fig. 3 that if the Pola-screen at the 




FIG. 4. Ray plane polarized by reflection. A ray of 
ordinary, unpolarized light is almost completely polarized 
when specularly reflected at about 32 to any non- 
metallic surface, such as glass. This permits eliminating 
oblique reflections from glass and water by a single Pola- 
screen over the lens. 

camera is rotated, some of the specular light will be allowed to pass 
through, so that the amount of specular reflection is under the con- 
trol of the photographer. When the camera Pola-screen is rotated 
so that its polarizing plane is actually parallel to that of the specular 
ray, this ray is transmitted even more freely than is the diffuse ray, 
so that the subject appears to have even brighter reflections and more 
gloss than it actually does have. 

Plane polarized light, or light that is partially plane polarized, is 
very common in nature, so that the photographer who is equipped 
with a Pola-screen only on his camera lens finds that he has rather 
considerable control over contrasts in his subject, even though he is 
unable to change the lighting of his subject. There are two sources 
of polarized light in nature: (1) Light rays from a clear blue sky, 
arriving at right angles to the sun's rays, are strongly polarized (see 



74 F. TUTTLE AND J. W. McFARLANE [J. S. M. P. E. 

Fig. 5) ; when this same skylight is specularly reflected from water, 
etc., these reflected rays are also polarized. (2) Ordinary, unpolar- 
ized light, specularly reflected from any non-metallic surface at about 
32 degrees to the surface, is strongly polarized by the act of reflection 
(see Fig. 4). There is some effect at other angles, but none at zero 
or 90 degrees. 

These two sources, separately and in combination, polarize much 
of the light from natural things. Unaided, our eyes do not detect 
polarized light, and so we have not seen until now that much of the 
light from our surroundings is polarized. Many natural things, 
seen through the Pola-screen, assume a new and strange beauty. 

These effects can be photographed easily with a Pola-screen over 




FIG. 5. Clear blue skylight, arriving at right angles 
to the sun's rays, is polarized. The sky may be darkened 
by the Pola-screen without affecting the color rendering 
of foreground objects. The strongest effect is attained 
with the camera axis roughly at right angles to the sun's 
rays. 

the lens alone. It will pay the photographer to view his subjects, 
both in the studio and outdoors, through a Pola-screen, while rotating 
it to see the effects. 



PROPERTIES OF THE POLA-SCREENS USEFUL IN MOTION PICTURE WORK 

A. Pola-Screen over Lens Alone. (1) Reducing polarized skylight 
to bring out clouds and other objects: A very dark sky may be ob- 
tained in color photography by this means which is impossible to 
achieve otherwise. The effect is greatest at right angles to the sun's 
rays. Therefore, at sunrise the region of greatest effect is north 
overhead south; at noon near the horizon in all directions; and 
at sunset north overhead south again. The arc swings from 



July, 1935] PHOTOGRAPHIC POSSIBILITIES OF POLARIZED LIGHT 75 

overhead to the west during the morning, from the east to overhead 
in the afternoon, passing through every part of the sky. 

Ordinary objects, faces, blossoms, trees, mountains, buildings, 
etc., can be made to stand out against the sky in a very beautiful 
manner. If desired, the brightness of the sky may be increased rela- 
tively to objects photographed against it, by rotating the Pola- 
screen to the appropriate position. In black-and-white photog- 
raphy, thePola-screen can serve as a filter of variable depth any- 
thing from red filter sky effects (without distortion of tone values) down 
to no filter effects may be attained by rotating the Pola-screen to 
the desired position. 

(2) Changing the contrast of various parts of a subject, without 
changing the lighting: This effect is very marked in the case of the 
walls and roofs of buildings, sunlighted water, and pavements from 
above. 

(3) Photographing subjects in water, from above: When the 
angle between lens axis and water surface is about 32 degrees, all re- 
flections from the water surface are eliminated. Reflections may be 
removed to some extent at other angles, but not at zero or 90 de- 
grees. 

Photography through glass or other transparent media: As in the 
case of water, at 32 degrees from the surface, reflections can be com- 
pletely removed. This effect can be used to produce double exposures 
by placing a thin pellicle mirror in front of the camera lens at the re- 
quired angle, and rotating the Pola-screen at the lens. The image 
reflected by the pellicle mirror appears or disappears according to the 
angle of the Pola-screen. Other more obvious applications will sug- 
gest themselves, such as photography of aquaria, through windows, 
and so on. 

B. P ola-Screens over Both Lens and Lights. (1) Subduing specular 
reflections from metallic and other glossy objects : Metallic reflections 
can not be eliminated entirely, but can be subdued very greatly. Re- 
flections from most other objects can be eliminated if desired. The 
Pola-screen over the camera lens is crossed with those over the lights 
for the greatest effect. 

(2) Increasing specular reflections: Articles may be made to ap- 
pear unnaturally glossy. The change, while considerable, is not as 
great as that possible in the opposite direction. The polarizing cells 
are used parallel. 

(5) Increasing color saturation: By removing the surface reflec- 



76 F. TUTTLE AND J. W. McFARLANE [J. S. M. P. E. 

tion, which is white, the colors of an object increase in their saturation, 
that is, their purity. The crossed arrangement produces such effects. 

(4) Effects upon faces: The crossed position produces a strange 
matte effect, with no luster whatever, and the facial colors are ex- 
aggerated. The parallel position has the opposite result a very per- 
spiry appearance, with the colors subdued. 

(5) Photographing wet objects: The surface reflections from wet 
objects, such as clinical specimens, present a severe problem, as they 
hide detail. These reflections may be subdued as desired, or elimi- 
nated at the crossed position. 

(6) Copying matte prints, pencil sketches, newsprint reproduc- 
tions, and paintings: Matte prints reflect light specularly in all di- 
rections. When this specular component is removed the blacks of 
the print become much blacker, so that the use of crossed Pola- 
screens produces a brightness range in the print that is even greater 
than that of a glossy print viewed in the normal manner. Likewise, 
the reflections can be removed from pencil graphite and ink particles, 
producing intense blacks. 

(7) Animation cells : Reflections from cells used in animation work 
build up with successive layers so that contrast is seriously affected, 
limiting the number of cells which may be used. These reflections 
may be greatly reduced by the crossed arrangement of Pola-screens. 

(8) Birefringent crystals and fibers: The phenomenon known as 
birefringence causes any transparent object, possessing the property, 
to light up, frequently in vivid color, when placed between two 
crossed Pola-screens. Cellophane, silk, cotton wool, and many 
natural crystals have this property. 

(9) Strained glass and celluloid : Any strained transparent medium 
displays birefringence, and when placed between crossed Pola- 
screens, shows a strain pattern. 

C. Applications in Lighting and Printing. The variable trans- 
mission of two Pola-screens together suggests a number of possibili- 
ties. Two Pola-screens used together over the lens constitute a 
variable neutral density filter, which may be of interest in making 
fades in some cases. The same arrangement can be used as an in- 
tensity control in a printer, and has the merits of simplicity; more- 
over, it does not cut down the area of the beam. 

Two light-source Pola-screens together can be used for controlling 
spotlight intensity. Cellophane added between these units intro- 
duces various color effects. 



July, 1935] PHOTOGRAPHIC POSSIBILITIES OF POLARIZED LIGHT 77 

Various lightings are possible with one Pola-screen over the lens 
and others at the lights. It is possible to place a back light so covered 
directly in the camera field. It is also possible to control the light 
reflected from any light so covered. A control, at the camera, is 
thus provided. It is therefore possible to photograph the same set 
with two cameras and obtain quite different lighting effects. 

TECHNICAL DATA 

The Eastman Pola-screens have a spectral range of polarizing 
power from 400-700mju. They absorb in the ultraviolet, and trans- 
mit freely, without polarization, in the infrared. They can be dam- 
aged by excessive heat, by placing them within a few inches of a lamp 
bulb, or imaging a lamp filament upon them. 

The most suitable negative materials are the panchromatic ma- 
terials now in general use. While it is possible to use the Pola- 
screens with orthochromatic or even with non-color-sensitized ma- 
terials, the exposure increase is very much greater. 

The exposure increase, for the Pola-screen over the lens alone, is 
about four times. For Pola-screens over both lens and lights, the 
exposure increase is ten times and upwards, depending upon the na- 
ture of the subject. When using a photoelectric type of exposure 
meter, the Pola-screen for the lens may be held over the meter win- 
dow at the intended angular position of the Pola-screen. The meter 
should always be used at the same angular position, as some of these 
meters are slightly polarizing in their sensitive element. 

The Pola-screens have a slight scattering power, so that those for 
lens use must be screened from all extraneous light by a proper lens 
hood. The Pola-screens supplied for light-source use are not suit- 
able optically for lens use. 

If calibrated angular scale is desired for repeating and recording 
settings used for Lens Pola-screens, the following is suggested: 



Parallel position 



Angle 


Scale Figure 


Effect 








Increased gloss 


45 


1 


Neutral 


60 


2 


Decreased gloss 


70 


3 


" 


76 


4 


ti 


80.5 


5 


(t 


84 


6 


<( 


86.5 


7 


n 



Crossed position 90 



78 F. TUTTLE AND J. W. McFARLANE 

The intervals of this scale will be of equal effectiveness in cutting 
down the polarized light entering the Pola-screen. 

The vibration plane of the Pola-screen meant for lens use is in line 
with the handle of the mount. 

The novelty of this subject makes it difficult to say just what ap- 
plication will be of most value in motion picture work. It is, how- 
ever, a new tool, by which new effects may be achieved, and its limita- 
tions are imposed only by the imagination of the user. We are in- 
debted to Mr. E. H. Land for help and suggestions, and to Dr. L. A. 
Jones for the demonstration film of polarization phenomenon in 
crystalline structure. 



DISCUSSION 

MR. DUNCAN: What is the crystallized material used in the filters? 

MR. CRABTREE: I do not know. It is a patented material. 

PRESIDENT TASKER: A discussion of the methods and materials is included 
in the U. S. Patent of E. H. Land, if you care to look that up. 

MR. DEPUE: When will the film be available? 

MR. CRABTREE : The author advised that a 2 1 /2-inch disk in a monocle mount 
will be available in B glass later in June, but the price is not yet determined. It 
will also be available in A glass, which should be used with long-focus lenses. 
The film itself will not be supplied for lens use because it is rippled and prevents 
good definition. It must be cemented between glass in order to get good 
definition. 

MR. JAMIESON: What is the difference of exposure with and without the filter 
in the Kodachrome process? 

MR. CRABTREE: The factor of one Pola-screen over the lens alone is 4, irre- 
spective of its angular position, for any panchromatic or color-film. When the 
light-sources are also covered by Pola-screens, the factor is 10 to 50, depending 
upon the subject. 

DR. FRAYNE : Is there a possibility of using these polarizing media in a den- 
silometer? 

MR. CRABTREE: It is physically possible, for a limited range. 

DR. FRAYNE: What is the cut-off? That is, what is the percentage of light 
transmitted from the darkest point? 

MR. CRABTREE: The optical density of two Pola-screens together, crossed, 
is 3; the transmission is therefore one-tenth of one per cent. 



A DEVICE FOR AUTOMATICALLY CONTROLLING THE 
BALANCE BETWEEN RECORDED SOUNDS* 

W. A. MUELLER** 



Summary. A re-recording method is described in which the dialog automatically 
controls the recorded level of the background effects or music, the dialog depressing 
the level of the effects or music from the normal value. This is accomplished by using 
the rectified dialog currents to control the grid-bias of a variable-mu tube. Its use 
results in greater intelligibility and realism in sound pictures. 



In making sound pictures, one of the most difficult problems is to 
maintain a proper balance between the dialog, music, and sound 
effects. If the music and effects are too loud, they mask the dialog 
and tend to make it unintelligible. When the music and effects are 
held low, so that the dialog will be intelligible, they are no longer 
lifelike and the picture loses its realism. This difficulty is also 
noticeable when the effects of large crowds, such as at a prize-fight, 
are combined with dialog. The two principals in the ring mutter 
threats to each other in a clinch. Their whispers, for the purpose 
of story continuity, must be heard and understood by everyone in 
the audience, even though ten thousand persons are yelling and 
stamping their feet at the same time. 

It has been standard practice to maintain the intelligibility of dialog 
during such a scene by holding the level of the effects low, or even 
eliminating them entirely. As a result, the scene lacks realism and 
much of its dramatic value is lost. If the effects are made loud 
enough to carry the realism of the scene, the dialog can not be under- 
stood, and the thread of the story is lost, while many in the audience 
turn to their neighbors to inquire what has been said. 

These difficulties were overcome some time ago by developing an 
automatic balance regulator by means of which the volume of the 
sound effects or music is varied automatically and simultaneously 
in accordance with the dialog level. The volume of the sound 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Warner Bros.-First National Studios, Burbank, Calif. 

79 



80 



W. A. MUELLER 



[J. S. M. P. E. 



effects is reduced as the dialog occurs, the extent of the reduction 
being in proportion to the level of the dialog. When the dialog 
ceases, the sound effects are restored to their normal value. 

A scene recorded in this manner gives a most pleasing result, as 
the sound effects are always real and lifelike, yet the slightest whisper 
is easily understood. This is possible only because the dialog in- 
stantly suppresses the volume of the effects, yet there is no delay 
between the end of the dialog and the restoration of the effects to 
normal volume. 

A block diagram of the device is shown in Fig. 1. The effects 
sound-tracks, which are to be combined with the dialog sound-track, 
are reproduced by machines 3, 4, and 5. The dialog sound-track 



D 




FIG. 1. Schematic diagram of automatic balance regulator. 

is reproduced by machine 10. The outputs of the various effects 
reproducers pass through the mixer attenuators shown, then into a 
variable-gain amplifier, 6, and from there through an attenuator into 
the common mixer bus, 12. 

The dialog reproduced in machine 10 passes through a mixer at- 
tenuator, an isolating amplifier, 11, another attenuator, and then to 
the common mixer bus, 12. Another amplifier, 13, is bridged across 
the output of the dialog reproducer for the purpose of supplying 
dialog to a rectifier, 14. The direct-current output of this rectifier 
controls the amplification of the variable-gain amplifier by changing 
the grid bias of the control grid of a variable-mu tube. As is well 
known, the effective gain of a variable-mu tube is reduced over a 
wide range as its grid bias is made more negative. Thus, the arrange- 
ment shown in Fig. 1 allows the dialog to control the regor^d volume 
of the effects automatically. 



July, 1935] BALANCE BETWEEN RECORDED SOUNDS 



81 



Fig. 2 is a diagram of the variable-gain amplifier and the associated 
rectifier controlling its gain. The effects or music to be controlled 
enter the variable-gain amplifier at its input, 75, and leave at its 




FIG. 2. Variable-gain amplifier and rectifier. 



output, 16. The gain of the amplifier is varied by changing the 
grid bias on the control grid of the variable-mu tube, 6. This is 
accomplished by the direct-current output of the rectifier tube, 




FIG. 3. Frequency characteristic of variable-gain amplifier. 



14, flowing through the potentiometer, 25. Thus, the gain of the 
variable-gain amplifier and, therefore, the level of the recorded 
effect are controlled by the level of the dialog impressed upon the 
input of the rectifier. 



S2 



W. A. MUELLER 



[J. S. M. P. E. 



It is essential that the variable-gain amplifier introduce no harmonic 
or frequency distortion and that it does not increase the noise level 
of the recording system. The frequency characteristic of this 
amplifier is shown in Fig. 3 for several values of grid bias and over-all 
gain. It is seen that the amplifier has a flat frequency characteristic 
which does not vary with changing grid bias. The present variable- 
mu tubes, with a suppressor grid, do not increase the noise level of 
the recording circuit by an amount that can be detected. 

This amplifier with an 
overload point of a 6 db., 
is installed in the recording 
circuit at the point of lowest 
level, the output that it is 
required to carry being only 
50 db. Consequently, the 
grid voltage variation on 
this tube is very small, and 
the possibility of harmonic 
distortion minimized. 
Numerous listening tests as 
well as harmonic analyses of 
single tones have conclusively 
proved that the introduction 
of this amplifier into the 
recording circuit does not 
impair the quality in the 
slightest degree. The extent 
to which the .variable-gain 
amplifier reduces the level 
of the effects or music is 
controlled by the resistance of the potentiometer, 25 . This control has 
been calibrated in 2-db. steps with a maximum gain reduction of 20 db., 
although 8 or 10 db. is all that it has been found necessary to use. 
To insure that the variation of the level of the effects will not be 
noticed even by the most expert listener, a delay circuit consisting 
of the two condensers, 36 and 48, and the inductance, 30, is placed 
in the output circuit of the rectifier. This delay circuit, which 
controls the rate at which the effects are varied, has been so ad- 
justed that for normal amounts of background reduction, the varia- 
tion in level of the effects is undetectable. 




INPUT TO RECTIFIER (v<X.T AC.) 

FIG. 4. Rectifier characteristic. 



July, 1935] BALANCE BETWEEN RECORDED SOUNDS 



83 



It has been found to be important that softly spoken dialog should 
depress the effects by the desired amount, but as the level of the 
dialog is increased, the maximum reduction is soon reached, after 
which there is no change in background level regardless of how loud 
the dialog becomes. This is accomplished by the resistor, 27 (Fig. 2), 
in the plate circuit of the rectifier, which limits the maximum value 
of rectified current as shown in the rectifier characteristic (Fig. 4). 

The rectifier characteristic of Fig. 4 results in the relationship 
between the dialog level and effects level shown in Fig. 5. Study of 
this curve shows that the reduction of the effects level is proportional 




20 25 30 

DIALOGUE LEVEL (0 B) 



FIG. 5. Level of dialog vs. level of effects for 13-db. reduction. 

to the dialog up to a certain value, beyond which the effects level 
does not change regardless of how loud the dialog becomes. 

The amplifier 13, Fig. 1, which is bridged across the dialog circuit 
and which feeds into the rectifier, is a three-stage amplifier. The 
gain of this amplifier drops off rapidly at low frequencies. This is 
necessary in order to prevent door slams, heavy footsteps, and other 
low-frequency sounds on the dialog track from varying the level of 
the effects. Consequently, the dialog frequencies have the maximum 
controlling effect. 

The isolating amplifier, 11, Fig. 1, is a high-quality one-stage 
amplifier with a flat frequency characteristic. The unilateral con- 
ductivity of this amplifier prevents the sound effects from the common 
mixer bus from being impressed upon the rectifier and interfering 



84 



W. A. MUELLER 



[j. s. M. p. E. 



with the control of the effects by the dialog channel. This amplifier 
is followed, as is shown in the block diagram, by an attenuator, 50. 
This attenuator properly terminates the amplifier, as well as allows 
its gain to be made equal to the normal gain of the variable-gain 
amplifier, which has been previously adjusted to zero by means of 
an attenuator. Consequently, the insertion of the background 
regulator into the recording circuit does not change any amplifier 
or mixer attenuator settings. 

The entire equipment is terminated in jacks, so that it can be 
patched into the circuit when required and removed when it is not 
needed. It is very flexible, as any number of effects sound-tracks 
can be controlled by the dialog, music, or any other sound. 




FIG. 6. Frequency characteristic of high-gain amplifier. 

The background regulator, as it is used, is very easy to operate 
and very practical. It gives a greatly improved product with the 
dialog perfectly understandable at all times, the background music 
loud and full, and effects that are loud enough to contribute to the 
realism of the picture. 

One of the most striking features of the device has been its ability 
to give an apparent increase in volume range. When the background 
effects are automatically reduced by a very loud sound on the dialog 
track, the effect on an observer is not that the background has been 
reduced in level, but that the desired sound has been tremendously 
increased in volume. This is a very desirable feature, as it enables 
certain sounds to be accentuated and greatly enhances their dramatic 
value. 



July, 1935] BALANCE BETWEEN RECORDED SOUNDS 85 

It is also very useful in smoothing out sequences in which some 
angles have background noises and other angles have been photo- 
graphed without any background effects. Considerable time and 
money are saved when large crowds are photographed, by dismissing 
the crowd as soon as the longer shots with the crowd and background 
effects are finished. The closer shots are made without the expense 
of the large crowd and with practically no background effects. Such 
a sequence when originally cut together is very unnatural, because 
in one cut the crowd will be heard, and in the next, there will be 
absolute silence. 

By using the background regulator to control an effects loop, 
whenever effects or dialog are present on the dialog track, the back- 
ground loop is suppressed. Whenever there are no effects or dialog 
on the dialog track, the background loop is recorded at full level. 
The result is a scene with a smooth background, resulting in a great 
improvement in the finished picture and a material saving in the 
cost of the production. 

DISCUSSION 

CHAIRMAN HANSEN: Warner Brothers have been working with this device 
for at least two or three years and probably have done more with it than any 
other studio. 

MR. KELLOGG: Perhaps it may have seemed so because our attention was 
called to it, but it appeared to me, at least, that the suppression of the back- 
ground noise was perhaps somewhat excessive. However, Mr. Mueller's state- 
ment that it has produced very desirable and pleasing effects is worth more in 
testimony than one's observations under such limited conditions as these, where 
we are all primed to expect an effect that is perhaps a little unnatural. 

I was interested in the explanation that the system employed here attempts 
practically to imitate by changes of relative intensity the psychological effect of 
switching attention from one sound to another. In actual life we can usually 
take advantage of differences of direction in order to concentrate attention upon 
a particular sound. The result of concentrating upon one sound is, of course, 
not to make the sound louder; but with our directive sense to help, we can 
largely forget the other sounds, which accomplishes the same purpose as making 
them actually fainter. Since, in the present case, all the sound comes from one 
direction, and our directive sense can not be brought into play, the suppression 
of the sounds in which the listener is less interested is accomplished by making 
them fainter. 

MR. MUELLER: In the demonstration reel I purposely over-accentuated the 
reduction of background effects so it would be very noticeable. The reduction 
was about twice the normal amount. I can see that we used enough so that 
everyone could see plainly how the device worked. 

CHAIRMAN HANSEN: I was wondering whether when the time arrives that 



86 W. A. MUELLER 

we have binaural recording at a point-source on the screen, we shall be able 
aurally to reject and keep the background level proper along with the dialog. 
What is the limiting factor of modulation with a musical background? At what 
frequency does the modulation effect start to come in? Where does the dialog 
modulate the music? 

MR. MUELLER: We have not made any tests on other than the present 
monaural recordings, but with the reduction normally used, which is about 8 
db., and with proper timing, the device works so well that we had to install the 
extension meter so that the mixers would know whether the device was in use or 
not. Under normal conditions, we have made a large number of pictures with 
it, and have never encountered the difficulty of its being noticed by anyone, 
even in our own department, nor of the public. 



HIGHLIGHTS OF 

THE SPRING CONVENTION 

HOLLYWOOD, CALIF., MAY 20-24, 1935 

HOTEL ROOSEVELT 

Throughout the week of the Convention probably as many as 400 or more 
persons attended from time to time, the attendance at the last session on Friday 
evening numbering nearly 350 persons. The Convention was probably the most 
outstanding, both as regards presentations and attendance, that the Society 
has ever held. The attendance from the East was 120. 

The meeting opened Monday morning with the usual Society business and 
the reports of the Membership and Convention Committees. A recommendation 
of the Historical Committee, ratified by the Board of Governors on the pre- 
ceding day, was presented to the Convention, to the effect that Mr. Thomas 
Armat, pioneer motion picture inventor now living in Washington, D. C., be 
admitted to Honorary Membership in the Society. With due procedure, the 
Convention approved the recommendation unanimously, and, accordingly, Mr. 
Armat will be awarded his Certificate of Honorary Membership at the Fall 
Convention, to be held at Washington, D. C., next October. 

At noon of the opening day, an informal luncheon was held for the members 
and guests. Short addresses of welcome were made by Mr. G. F. Rackett, on 
behalf of the Pacific Coast Section, and by Mr. E. Huse, Executive V ice-President 
of the Society, followed by a brief response by President Tasker. Short ad- 
dresses followed, by Mr. Howard Green, writer, Paramount Productions, Inc.; 
Mr. Kenneth Macgowan, Associate Producer, RKO Radio Productions, Inc.; 
and Mr. George E. Browne, International President, I. A. T. S. E. (due to illness, 
Mr. Browne was unable to attend; his address was read by Mr. Thad Barrows). 

The program of papers and presentations, as actually followed at the sessions, 
was as published on succeeding pages of this issue of the JOURNAL. On Monday 
evening, Mr. W. Garity, Production Supervisor of Walt Disney Studios, enter- 
tained approximately 150 of the members with a demonstration of some of the 
means employed for creating the incidental sounds for the Mickey Mouse cartoons 
and the Silly Symphonies, followed by a formal paper on the subject of cartoon 
technic. The meeting concluded with the projection of several unfinished 
cartoons by way of explication. It was interesting to learn that the animation 
is now adapted to the music and not vice versa. 

On Tuesday, at noon, the members were entertained by the Electrical De- 
partment of Warner Bros. -First National at a luncheon in the Studio restaurant, 
under the direction of Mr. Frank Murphy, Chief Studio Engineer. After the 
luncheon, at which President Tasker briefly addressed the gathering, the party 
spent the remainder of the afternoon touring the studios. 

Tuesday evening, the members of the S. M. P. E. were the guests of the Academy 
of Motion Picture Arts and Sciences at a meeting of the Technicians Branch, 

87 



88 HIGHLIGHTS OF SPRING CONVENTION [J. S. M. p. E. 

held at the Carthay Circle Theater, in Hollywood. Papers were presented by 
J. A. Ball, Mrs. N. Kalmus, and R. Mamoulian, on the subject of color in motion 
pictures (see program on succeeding pages), followed by various examples of 
recent color motion pictures, including a reel of the currently released feature 
Becky Sharp. Major N. Levinson, Chairman, Technicians Branch of the Acad- 
emy, presided, with Mr. K. Macgowan as guest chairman. 

On Wednesday afternoon the members were conducted on a tour through the 
beautiful Fox Hill Studios of the Fox Film Corp., under the direction of Mr. W. J. 
Quinlan, Chief Studio Engineer. 

At the Semi-Annual Banquet and Dance, held in the New Supper Room of the 
Hotel Roosevelt on Wednesday evening, the members were addressed by Mr. 
Frank Lloyd, of Metro-Goldwyn-Mayer and President of the Academy of Motion 
Picture Arts and Sciences, after a brief introduction by President Tasker. Direc- 
tion of the proceedings was then taken over by Bill Ray of Warner Bros. Radio 
Station KFWB, who acted as master of ceremonies, introducing the various 
motion picture stars and celebrities who entertained the gathering before the 
microphone. The proceedings were broadcast through the courtesy of Mr. Jerry 
King over KFWB and associated stations of the Southern California Network. 

On Thursday afternoon a trip was arranged to the California Institute of 
Technology, for those who did not attend the technical session that afternoon. 
A group of about fifty persons was conducted through the aeronautic, cosmic 
ray, and high-voltage laboratories under the direction of Dean F. W. Hinrichs, Jr. 

THE PAPERS PROGRAM 

In spite of the large number of papers on the program, almost without exception 
these were run strictly according to schedule, thanks to the masterly job of 
presiding by President Tasker. The standard of quality of the papers was 
generally of a high order and a number of outstanding papers may be mentioned 
as follows: 

"The Kodachrome Process of Amateur Cinematography in Natural Colors" 
was described by L. Mannes and L. Godowsky, and a reel of color film pro- 
jected. This process is epoch-making in so far as the film can be exposed in a 
motion picture camera without the use of auxiliary devices and, after processing, 
a three-color image is obtained, the graininess of which is no greater than that of 
the average silver image. 

F. W. Tuttle and J. W. McFarlane, in their paper entitled "Introduction to the 
Photographic Possibilities of Polarized Light," announced the availability of 
polarizing filters in sheet form which consist essentially of crystals of certain 
chemical compounds oriented in a flexible film base. By rotating two such 
filters into the "crossed" position the incident light is effectually extinguished. 
It was pointed out that reflected light, in the form of glare, is largely plane 
polarized, so that by placing a single polarizing filter over the lens of the camera 
and suitably rotating it, the glare can be effectually eliminated. A demonstra- 
tion film was shown, illustrating how the glare from automobiles, reflections 
when photographing store windows, etc., can be eliminated completely in this 
manner. In color photography, a blue sky can likewise be darkened without 
affecting the brightness of the foreground. By placing polarizing filters over 



July, 1935] HIGHLIGHTS OF SPRING CONVENTION 89 

the lamps and, likewise, over the lens, interesting special effects can be obtained. 
These filters would appear to be a useful tool for the cameraman. 

R. R. Scoville described instruments for measuring flutter, and T. E. Shea, 
W. A. MacNair, and V. Subrizi explained the effect of flutter on sound quality 
and gave an enlightening demonstration. 

At the symposium on color photography arranged by the Technicians Branch 
of the Academy of Motion Picture Arts and Sciences the potential importance 
of color to the motion picture industry was demonstrated forcibly and, from the 
masterly extemporaneous talk by Mr. Rouben Mamoulian on "Some Problems 
in Directing Color Motion Pictures," it is apparent that the industry will adopt 
color almost universally just as soon as a color process is available which is 
sufficiently cheap, gives correct color rendering and good definition, does not 
require excessive studio illumination, and can be operated by existing studio 
personnel. 

The largest attendance throughout the convention was during the Sound 
Session on Friday evening which included three epoch-making demonstrations. 
The first was that of Douglas Shearer of a push-pull method of recording, in- 
volving the use of as many as four light-valve ribbons. The recordings were 
reproduced through a 60-watt amplifier, using special horns from the Bell Tele- 
phone Laboratories. The quality of the reproduced sound was unquestionably 
the best that has ever been reproduced before our Society, and the resulting 
sensation simulated very closely that from a large symphonic orchestra. 

M. C. Batsel repeated his demonstration of high-quality reproduction first 
given at the Atlantic City meeting of the Society a year ago. 

J. A. Miller described the "Mechanographic Recording of Motion Picture 
Sound-Tracks. " This consists in recording with a F-shaped stylus upon a special 
motion picture film which, apparently, consists of film-base coated with a thick 
gelatin coating upon which is superimposed a very thin opaque layer. As the 
stylus vibrates, the opaque layer is removed and a twin variable-width track is 
produced. In spite of inadequate reproducing equipment, the quality of the 
reproduced sound was such as to indicate the potential value of this method of 
recording, especially for play -backs. 

THE APPARATUS EXHIBIT 

The large attendance indicated appreciation of our members of this feature of 
the convention. Some of the equipment included in the exhibit was described 
in the Apparatus Symposium listed in the detailed program on the following 
pages, and will, of course, be described in the JOURNAL. The following firms 
exhibited their new equipment. 

Ampro, Inc. Eastman Kodak Co. 

Ashcraft Mfg. Co. Electrical Research Products, Inc. 

Baldor Electric Co. Electro-Acoustics Products Co. 

Cannon Electric Development Co. General Electric Co. 

Century Electric Co. Goldberg Bros. 

O. B. Depue Hollywood Camera Exchange 

DeVry Sound System Mole-Richardson, Inc. 

Dictograph Products Co. Moviola Company 



90 HIGHLIGHTS OF SPRING CONVENTION 

National Carbon Co. Newmann Process Projector Co. 

National Theater Supply Co. RCA Manufacturing Co. 

Neumade Products Corp. SCK Corporation 

Vitachrome, Inc. 

ACKNOWLEDGMENT 

Credit for the success of the Convention, which may be measured in terms of 
great increase of interest in the activities of the Society, its Conventions and 
Local Section meetings, its JOURNAL and technical activities, was largely due to 
the efforts of Mr. W. C. Kunzmann, Convention Vice-President; Mr. J. I. 
Crabtree, Editorial Vice-P resident; Mr. J. O. Baker, Chairman of the Papers 
Committee; Mr. W. E. Mueller, Vice-Chairman of the Papers Committee; Mr. 
H. Griffin, in charge of projection; Mr. P. Mole, Chairman, Local Arrangements 
Committee; Mr. Ted Lay and his staff, for installing the equipment; Mr. O. F. 
Neu, in charge of the Apparatus Exhibit; Mrs. E. Huse, hostess, in charge of the 
ladies' activities; and the officers and members of the Los Angeles Local 150 
I. A. T. S. E. 

The Chairman and Board of Managers of the Pacific Coast Section of the 
Society especially are to be thanked for their untiring efforts and valuable guidance 
toward making the Convention a success. 

Others to whom credit is due were Mr. W. Garity, Mr. F. Murphy, and Mr. 
W. J. Quinlan, for arranging the visits to the Walt Disney, Warner Bros.-First 
National, and Fox Hill Studios, respectively. Thanks are due to the manage- 
ment of Grauman's Chinese and Egyptian Theaters, Pantages' Hollywood Theater, 
Warner Bros.' Hollywood Theater, and Gore Bros.' Iris Theater for the passes 
courteously supplied to the members during the week of the Convention. Mr. 
Carl Schaefer, of the Warner Bros. Publicity Department, is to be thanked and 
congratulated for his splendid publicity work. 

The sound and projection equipment used at the meetings was supplied and 
installed by the International Projector Corp., the National Theater Supply Co., 
Bausch & Lomb Optical Co., National Carbon Co., Raven Screen Co., Electrical 
Research Products, Inc., RCA Manufacturing Co., Mole-Richardson, Inc. 

Acknowledgment is due to the Academy of Motion Picture Arts and Sciences, 
particularly Mr. Gordon S. Mitchell and Major N. Levinson, for their assistance 
in arranging various functions and presentations ; for the dinner tendered to the 
Board of Governors on Tuesday, at the Victor Hugo Cafe in Beverley Hills; 
for the invitation to the Technicians Branch meeting on Tuesday evening; and 
for the general clerical and other assistance rendered by the Academy. 

Kind assistance was also rendered by the J. Slipper Company in connection 
with the Apparatus Exhibit; and the Royal Typewriter Company, Hollywood 
Branch, in connection with the registration activities. 

Prints of the photograph taken at the banquet on Wednesday evening may be 
obtained for one dollar each from the Weaver Photo Service, 1041 W. 42nd Place, 
Los Angeles. Members who were in the photograph taken at the Fox Hill 
Studio, and who have not yet received their prints, may obtain a print free of 
charge by writing to the General Office of the Society. 



PROGRAM OF THE 
SPRING CONVENTION AT HOLLYWOOD, CALIF. 

MAY 20-24, 1935; HOTEL ROOSEVELT 
MONDAY, MAY 20th 

10:00 a.m. General Session, Gerald F. Rackett, Chairman, Pacific Coast Sec- 
tion of the S. M. P. E. Presiding. 

Society Business. 

Report of the Membership and Subscription Committee, E. R. Geib, Chairman. 

Report of the Progress Committee, J. G. Frayne, Chairman. 

"Television and Motion Pictures"; A. N. Goldsmith, New York, N. Y. 

"Theatrical Possibilities of Television"; H. R. Lubcke, Don Lee Broad- 
casting System, Hollywood, Calif. 

Report of the Historical Committee, W. E. Theisen, Chairman. 

"The Talking Book"; J. O. Kleber and L. Thompson, American Foundation 
for the Blind, New York, N. Y. 

"Use of Films and Motion Picture Equipment in Schools"; Miss M. Evans. 
San Diego City Schools, San Diego, Calif. 

12:30 p.m. Informal Get-Together Luncheon. 

Addresses of Welcome: On behalf of the Pacific Coast Section, S. M. P. E., 
Emery Huse, Executive Vice- President, S. M. P. E. On behalf of the 
Academy of Motion Picture Arts and Sciences, Major N. Levinson, Chair- 
man, Technicians Branch and Vice-Chairman, Research Council, Academy 
of Motion Picture Arts and Sciences. 

Response: Homer G. Tasker, President, Society of Motion Picture Engineers. 
Addresses by 

Howard Green, Writer, Paramount Productions, Inc., Hollywood, Calif. 
Kenneth Macgowan, Associate Producer, RKO Radio Productions, Holly- 
wood, Calif. 

George E. Browne, International President, I. A. T. S. E. and M. P. M. O. U., 
Washington, D. C. 

2:00 p.m. General Session. Homer G. Tasker, President S. M. P. E. Pre- 
siding. 

"A Description of the Historical Motion Picture Exhibit in the Los Angeles 

Museum"; W. E. Theisen, Honorary Curator, Motion Picture and Theatrical 

Arts Section, Los Angeles Museum, Los Angeles, Calif. 
"Production Problems of the Writer Related to the Technician"; C. Wilson, 

Metro-Goldwyn-Mayer Studios, Culver City, Calif. 
"The Kodachrome Process of Amateur Cinematography in Natural Color"; 

L. Mannes and L. Godowsky, Eastman Kodak Company, Rochester, N. Y. 

91 



92 PROGRAM OF SPRING CONVENTION [j. S. M. p. E. 

"The Inter-Relation of the Dramatic and Technical Aspects of Motion Pic- 
tures"; Prof. B. V. Morkovin, University of Southern California, Los 
Angeles, Calif. 

"Introduction to the Photographic Possibilities of Polarized Light"; F. W. 
Tuttle and J. W. McFarlane, Eastman Kodak Company, Rochester, N. Y. 

"The Problems of a Motion Picture Research Library"; Miss H. G. Percey, 
Paramount Productions, Inc., Hollywood, Calif. 

8:30 p.m. Studio Visit. 

Visit to Walt Disney Studio, under the direction of W. Garity, Production 
Supervisor. 

TUESDAY, MAY 21st 

9:30 a.m. Studio Session. Douglas Shearer, Metro-Goldwyn-Mayer, Pre- 
siding. 

Report of the Committee on Standards and Nomenclature, E. K. Carver, 
Chairman. 

"Flutter in Sound Records"; T. E. Shea, W. A. MacNair, and V. Subrizi, Bell 
Telephone Laboratories, Inc., New York, N. Y. 

"Portable Flutter Measuring Instruments"; R. R. Scoville, Electrical Re- 
search Products, Inc., Hollywood, Calif. 

"Some Background Considerations of Sound System Service"; J. S. Ward, 
Electrical Research Products, Inc., New York, N. Y. 

"Modern Methods of Servicing Sound Motion Picture Equipment"; C. C. 
Aiken, RCA Manufacturing Company, Camden, N. J. 

"Technic of Present- Day Motion Picture Photography"; V. E. Miller, Para- 
mount Studios, Hollywood, Calif. 

"Engineering Technic in Pre-Editing Motion Pictures"; M. J. Abbott, RKO 
Studios, Hollywood, Calif. 

"The Analysis of Harmonic Distortion in a Photographic Sound Record by 
Means of an Electrical Frequency Analyzer"; O. Sandvik, V. C. Hall, and 
W. K. Grimwood, Eastman Kodak Company, Rochester, N. Y. 

1 :30 p.m. Luncheon and Studio Visit. 

Luncheon on the lot, and' inspection of Warner Bros.-First National Studio; 
courtesy of the Electrical Department, under the direction of F. Murphy, 
Chief Studio Engineer. 

8:30 p.m. Meeting of the Technicians Branch of the Academy of Motion 

Picture Arts and Sciences; Carthay Circle Theater, Hollywood. Major N. 
Levinson, Chairman, Technicians Branch and Vice-Chairman, Research 
Council of the Academy, Presiding. Kenneth Macgowan, Guest Chairman. 

"The Technicolor Process"; J. A. Ball, Technicolor Motion Picture Corpora- 
tion, Hollywood, Calif. 

"Psychology of Color"; Natalie Kalmus, Technicolor Motion Picture Cor- 
poration, Hollywood, Calif. 

"Some Problems in Directing Color Motion Pictures"; Rouben Mamoulian 
Director, Hollywood, Calif. 



July, 1935] PROGRAM OF SPRING CONVENTION 93 

WEDNESDAY, MAY 22nd 

9:30 a.m. Laboratory Session. Emery Huse, Executive Vice-President, 
S. M. P. E., Presiding. 

"The Argentometer an Apparatus for Testing for Silver in a Fixing Bath"; 

W. Weyerts and K. C. D. Hickman, Eastman Kodak Company, Rochester, 

N. Y. 
"Motion Picture Film Processing Laboratories in Great Britain"; I. D. 

Wratten, Kodak Limited, London, England. 

"Optical Printing and Technic"; Lynn Dunn, RKO Studios, Hollywood, Calif. 
"A Continuous Printer for Optically Reducing a Sound Record from 35-Mm. 

to 16-Mm. Film"; O. Sandvik and J. G. Streiffert, Eastman Kodak Com- 
pany, Rochester, N. Y. 
"Non-Uniformity in Photographic Development"; J. Crabtree, Bell Telephone 

Laboratories, Inc., New York, N. Y. 
"A Dynamic Check on the Processing of Film for Sound Records"; F. G. 

Albin, United Artists Studios, Hollywood, Calif. 
"Emulsions for Special Fields in Motion Picture Photography"; W. Leahy 

Agfa Ansco Corporation, Hollywood, Calif. 
"Sensitometric Studies of Processing Conditions for Motion Picture Film"; 

H. Meyer, Agfa Ansco Corporation, Hollywood, Calif. 

2:30 p.m. Studio Visit. 

A Visit to the Fox Hill Studio, under the direction of W. J. Quinlan, Chief 
Studio Engineer. 

7:30 p.m. Semi- Annual S. M. P. E. Banquet. 

The semi-annual banquet and dance of the Society was held in the New Supper 
Room of the Hotel. Addresses by Frank Lloyd, Director, M-G-M, and 
President, Academy of Motion Picture Arts and Sciences; Rouben Mamou- 
lian, Director; star presentations; broadcast through Warner Bros.' Radio 
Station, KFWB, and associated stations of the Southern California Network. 

THURSDAY, MAY 23rd 

9:30 a.m. Projection and Studio Lighting Session. Hollis W. Moyse, Dupont 

Film Mfg. Corp., Presiding. 

Report of the Projection Practice Committee, J. O. Baker, Chairman. 
Report of the Projection Screen Brightness Committee, C. Tuttle, Chairman. 
Report of Non-Theatrical Equipment Committee, R. F. Mitchell, Chairman. 
"Non-Theatrical Projection"; R. F. Mitchell, Bell & Howell Company, 

Chicago, 111. 
"The Relation between Projector Illumination and Screen Size for Non- 

Theatrical Projection," D. Lyman, Eastman Kodak Company, Rochester, 

N. Y. 
"Sixteen-Mm. Negative-Positive and Grain"; D. Norwood, Lt., U. S. Army 

Air Corps, Chanute Field, Rantoul, 111. 
"Trends in Sixteen-Mm. Projection with Special Reference to Sound"; A, 

Shapiro, Ampro Corporation, Chicago, IU, 






94 PROGRAM OF SPRANG CONVENTION [j. S. M. p. E. 

Report of the Studio Lighting Committee, R. E. Farnham, Chairman. 

"The Radiant Energy Delivered on Motion Picture Sets from Carbon Arc 
Studio Light Sources"; F. T. Bowditch and A. C. Downes, National Car- 
bon Company, Cleveland, Ohio. 

"The Photographic Effectiveness of Carbon Arc Studio Light Sources"; F. T. 
Bowditch and A. C. Downes, National Carbon Company, Cleveland, 
Ohio. 

"Lighting for Technicolor Motion Pictures"; C. W. Handley, National Car- 
bon Company, Los Angeles, Calif. 

"A New Wide-Range Spot Lamp"; E. C. Richardson, Mole-Richardson, Inc., 
Hollywood, Calif. 

"Sources of Direct Current for Non-Rotating High-Intensity Reflecting Arc 
Lamps"; C. C. Dash, Hertner Electric Company, Cleveland, Ohio. 

2:00 p.m. Sound and Standardization Session. E. H. Hansen, Fox Film 
Corp., Presiding. 

"The Technical Aspects of Recording Music for Motion Pictures"; R. H. 
Townsend, Fox Film Company, Hollywood, Calif. 

"Pioneering in Motion Pictures"; Dr. Lee deForest, Hollywood, Calif. 

"A Device for Automatically Controlling the Balance between Recorded 
Sounds"; W. A. Mueller, Warner Bros.-First National, Burbank, Calif. 

"Improvements in Play-Back Disk Recording"; G. M. Best, Warner Bros.- 
First National, Burbank, Calif. 

"Process Cinematography"; J. A. Norling, Loucks and Norling, New York, 
N. Y. 

2:30 p.m. California Institute of Technology. 

A visit to the Institute, under the direction of Dean F. W. Hinrichs, Jr.; 
inspection of the astronomical, aeronautic, cosmic ray, and high-voltage 
laboratories. 

8:00 p.m. Studio Session, John G. Frayne, Electrical Research Products, Inc., 
Presiding. 

Report of the Sound Committee, P. H. Evans, Chairman. 

"Improvements in Sound Quality of Newsreels"; J. A. Battle, Electrical Re- 
search Products, Inc., New York, N. Y. 

"Analysis of the Distortion Resulting from Sprocket-Hole Modulation"; 
E. W. Kellogg, RCA Manufacturing Company, Camden, N. J. 

"Wide-Range Reproduction in Theaters"; J. P. Maxfield and C. Flannagan, 
Electrical Research Products, Inc., New York, N. Y. 

"Characteristics of the Photophone Light-Modulating System"; L. T. Sachtle- 
ben, RCA Manufacturing Company, Camden, N. J. 

"The Standardization of Make-Up"; M. Factor, Max Factor, Inc., Holly- 
wood, Calif. 

FRIDAY, MAY 24th 

9:30 a.m. Sound and Acoustics Session. Kenneth F. Morgan, Electrical Re- 
search Products, Inc., Presiding. 



July, 1935] PROGRAM OF SPRING CONVENTION 95 

"Modern Instruments for Acoustical Studies"; E. C. Wente, Bell Telephone 

Laboratories, New York, N. Y. 
"Recent Developments in Architectural Acoustics"; V. O. Knudsen, Professor 

of Physics and Dean of Graduate Study, University of California at Los 

Angeles, Calif. 

"Principles of Measurements of Room Acoustics"; E. C. Wente, Bell Tele- 
phone Laboratories, New York, N. Y. 
"Studio Acoustics"; M. Rettinger, Pacific Insulation Company, Los Angeles, 

Calif. 
"The Technical Aspects of the High-Fidelity Reproducer"; E. D. Cook, RCA 

Manufacturing Company, Camden, N. J. 
"Development and Design of the High-Fidelity Reproducer"; F. J. Loomis and 

E. W. Reynolds, RCA Manufacturing Company, Camden, N. J. 
"Calibrated Multi-Frequency Test Film"; F. C. Gilbert, Electrical Research 

Products, Inc., New York, N. Y. 

2:00 p.m. General Session. Joseph A. Dubray, Bell & Howell Co., Presiding. 

"Technical Aspects of the Motion Picture" ; A. N. Goldsmith, New York, N. Y. 
"The History of the Talking Picture"; W. E. Theisen, Hollywood, Calif. 

Apparatus Symposium. 

"Three New Kodascopes"; N. Green, Eastman Kodak Company, Rochester, 

N. Y. 
"A Continuous Film Camera for High-Speed Photography"; C. T. Burke, 

General Radio Company, Cambridge, Mass. 
"A Professional 16-Mm. Projector with Intermittent Sprocket"; H. A. DeVry, 

Herman A. DeVry, Inc. Chicago, 111. 
"Arc Supply Generator for Use with Suprex Carbons"; W. K. Hartman, 

Century Electric Company, Los Angeles, Calif. 
"A Sound Reduction Printer"; O. B. Depue, Chicago, 111. 
"A 35-Mm. Automatic Daylight Sound Motion Picture Projector"; A. B. 

Scott, SCK Corporation, Hollywood, Calif. 
"Vitachrome Diffusionlite System and Lamps, Their Uses and Applications"; 

A. C. Jenking, Vitachrome, Inc., Los Angeles, Calif. 
"The Cinemaphone Unit Cabinet for Reproducing 16-Mm. Sound Pictures"; 

F. J. Hawkins, Los Angeles, Calif. 
"The Edmison Film Protective Device for Preventing Ignition of Film during 

Projection"; F. J. Hawkins, Los Angeles, Calif. 
"A New Sound Reader and Frame Viewer"; I. Serrurier, Moviola Co., 

Hollywood, Calif. 
"The New Wall Sound Camera"; H. Griffin, International Projector Corp., 

New York, N. Y. 

"A New Background Projector for Process Cinematography"; H. Griffin, In- 
ternational Projector Corp., New York, N. Y. 

"The Use of Cinematography in Aircraft Flight Testing"; F. H. Collbohm, 

Douglas Aircraft Company, Inc., Santa Monica, Calif. 
"The Use of Motion Pictures for Human Power Measurements"; J. M. Albert, 

Chas. E. Bedaux Company, San Francisco, Calif. 



96 PROGRAM OF SPRING CONVENTION 

"The Motion Picture in Japan"; Y. Osawa, J. Osawa and Company, Ltd., 

Kyoto, Japan. 
"The Motion Picture Industry in India"; G. D. Lai, Delhi, India. 

8:00 p.m. Sound Session. Homer G. Tasker, President, S. M. P. E., Pre- 
siding. 
"A Variable-Density Recording Method to Produce Increased Undistorted 

Volume Range"; Douglas Shearer, Metro-Goldwyn-Mayer Studios, Culver 

City, Calif. 
"Recording Music for Motion Pictures"; M. C. Batsel, RCA Manufacturing 

Company, Camden, N. J. 
"Mechanographic Recording of Motion Picture Sound-Track"; J. A. Miller, 

Miller Film, Inc., New York, N. Y. 
"A Comparison of Variable- Density and Variable- Width Sound Records"; 

E. W. Kellogg, RCA Manufacturing Company, Camden, N. J. 
"The S. M. P. E. Progress Medal"; H. G. Tasker, President, S. M. P. E. 
"A Consideration of Some Special Methods of Re-Recording"; E. D. Cook, 

RCA Manufacturing Company, Camden, N. J. 



SOCIETY ANNOUNCEMENTS 

FALL, 1935, CONVENTION 

The next Convention of the Society will be held at Washington, D. C., October 
21-24th, inclusive, headquarters at the Wardman Park Hotel. Members are urged 
to keep the dates in mind and to plan ahead, so as to assure a good attendance. 
W. C. Kunzmann, Convention V ice-President, is already engaged in arranging the 
details of the Convention; and the Papers Committee, under the direction of J. I. 
Crabtree, Editorial Vice-President, and J. O. Baker, Chairman of the Committee, 
will soon begin to prepare the technical program of papers and presentations. 

SECTIONAL COMMITTEE ON MOTION PICTURES UNDER THE ASA 

Following the last meeting of the Committee, letter-ballots were mailed to all 
the members for voting upon the 16-mm. film and equipment standards contained 
in the SMPE Standards Booklet and published in the Nov., 1934, issue of the 
JOURNAL. The approval of the standards as thus published was unanimous; 
whereupon the project was referred to the SMPE Board of Governors for valida- 
tion as sponsor and for subsequent transmittal to the ASA for adoption as Ameri- 
can national standards. 

At the meeting of the Board of Governors at Hollywood on May 19th, the Board 
ratified the previous action of the SMPE Standards Committee, and adopted 
all the specifications contained in the Standards Booklet as SMPE Standards. 
It ratified also the action of the Sectional Committee in respect to the 16-mm. 
standards, and drafted a letter of transmittal in that regard to the American 
Standards Association, together with a list of the personnel approved by the 
Board as sponsor. The list is identical to the list published in the April, 1935, 
issue of the JOURNAL, p. 378, excepting that D. B. Joy now represents the Na- 
tional Carbon Company; the American Society of Cinematographers is repre- 
sented by G. A. Mitchell; and the SMPE by A. N. Goldsmith, E. K. Carver, and 
H. G. Tasker. A. N. Goldsmith is the chairman of the Sectional Committee. 

SIXTEEN-MM. SOUND-FILM STANDARDIZATION 

At the National Film Congress (Reichsfilmkammer) held at Berlin on April 
25th, G. Friedl represented the SMPE and the Sectional Committee in connection 
with the discussion of 16-mm. sound-film standardization. Unfortunately, the 
final action of the Congress was to ratify the findings of the previous conference 
at Stresa, which included, among other things, placing the sound-track at the 
opposite edge of the film from that specified in the SMPE standards. This 
means, then, that unless the DIN standards are changed at the forthcoming con- 
ference at Paris on July 7th under the auspices of the International Congress of 
Scientific and Applied Photography and the International Standards Association, 

97 



98 SOCIETY ANNOUNCEMENTS [J. S. M. p. E. 

there will be two different 16-mm. sound-film standards in Europe, the SMPE 
and the one proposed by the International Institute of Educational Cinema- 
tography (ICE) and ratified at Berlin by the Reichsfilmkammer (DIN = Deut- 
schen Industrie Normen). 

A presentation for the Paris conference is now being prepared, which will be 
published shortly in the JOURNAL. Mr. Friedl will again represent the American 
interests. 

NOMINATIONS FOR OFFICERS AND GOVERNORS FOR 1936 

Ballots for nominating officers and governors of the Society for 1936, to take 
the places of those whose terms expire at the end of this year, were recently mailed 
to the voting membership of the Society. The ballots, when returned, will form 
the basis of the nominations made by the Board of Governors at its next meeting 
at New York on July 19th. 

Voting ballots will subsequently be mailed to the Fellow and Active members 
of the Society, they will be counted at the Fall Convention at Washington, Oct. 
21st, and the successful candidates will assume their offices on Jan. 1st. The 
officers and governors whose terms expire Dec. 31, 1935, are as follows: 

H. G. TASKER, President T. E. SHEA, Treasurer 

E. HUSE, Executive V-P. J. H. KURLANDER, Secretary 

L. A. JONES, Engineering V-P. A. S. DICKINSON, Governor 

O. M. GLUNT, Financial V-P. H. GRIFFIN, Governor 

W. B. RAYTON, Governor 

Nominations and voting for officers and Boards of Managers of the three Local 
Sections are likewise being conducted concurrently, the results of which will be 
announced later. 

PROGRESS MEDAL AWARD 

Action is being taken by the Committee on the Progress Medal Award to select 
the recipient of the medal for the year 1934. The recommendation of the Com- 
mittee will be presented to the Board of Governors on July 19th, and the presenta- 
tion will be made at the Fall Convention at Washington. Names of persons 
deemed worthy of the award may be proposed and seconded in writing by any 
two Fellows or Active members of the Society. A written statement of the ac- 
complishments of the nominee should accompany each proposal. 

The Progress Medal was designed by Mr. Alexander Murray of Rochester. 
In recognition of the excellence of his work, the Board of Governors at its last 
meeting awarded to Mr. Murray a bronze replica of the medal. 

ERRATUM 

The following change should be made in the paper entitled "My Part in the 
Development of the Motion Picture Projector" by Thomas Armat, in the March, 
1935, issue of the JOURNAL: 

In the second paragraph on p. 243, the patent number 536,539 should read 
536,569. 



July, 1935] SOCIETY ANNOUNCEMENTS 99 



The Society regrets to announce the death of 

Eugene Augustin Lauste 

Honorary Member of the Society 

June 27, 1935 



SOCIETY SUPPLIES 

Reprints of Standards of the SMPE and Recommended Practice may be obtained 
from the General Office of the Society at the price of twenty-five cents each. 

Copies of Aims and Accomplishments, an index of the Transactions from Octo- 
ber, 1916, to June, 1930, containing summaries of all the articles, and author and 
classified indexes, may be obtained from the General Office at the price of one 
dollar each. Only a limited number of copies remains. 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 
JOURNAL, may be obtained from the General office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



STANDARD S. M. P. E. 

VISUAL AND SOUND TEST REELS 

Prepared under the Supervision 

OF THE 
PROJECTION PRACTICE COMMITTEE 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 
Reels sold complete only (no short sections). 

PRICE $37.50 FOR EACH SECTION, 
INCLUDING INSTRUCTIONS 

(Shipped to any point in the United States) 

Address the 

SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL PENNSYLVANIA 
NEW YORK, N. Y. 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV AUGUST, 1935 Number 2 



CONTENTS 

Page 

Recording Music for Motion Pictures M. G. BATSEL 103 

Improvements in Playback Disk Recording G. M. BEST 109 

A Continuous Optical Reduction Sound Printer 

O. SANDVIK AND J. G. STREIFFERT 117 

The Technicolor Process of Three- Color Cinematography 

J. A. BALL 127 

Color Consciousness NATALIE M. KALMUS 139 

Some Problems in Directing Color Pictures R. MAMOULIAN 148 

Improvements in Sound Quality of Newsreels. . . . J. A. BATTLE 154 

A Dynamic Check on the Processing of Film for Sound Records 

F. G. ALBIN 161 

Engineering Technic in Pre-Editing Motion Pictures 

M. J. ABBOTT 171 

Characteristics of Photophone Light-Modulating System 

L. T. SACHTLEBEN 175 

Report of the Standards Committee 192 

Fall Convention: October 21-24, inclusive, Washington. D.C. 195 

Society Announcements 198 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York. N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



RECORDING MUSIC FOR MOTION PICTURES 1 



M. C. BATSEL** 

Summary. The performance standards required for musical reproduction were 
established in the research laboratories of the General Electric Company early in the 
period of development of Photophone equipment. Experience has proved the neces- 
sity of the refinements found desirable in the research laboratories. 

In this paper several developments are referred to that have reduced distortion. It 
is pointed out that perfect equipment does not necessarily result in satisfactory musical 
reproduction. Consideration must be given to the acoustical properties of scoring 
stages. 

The initial ideals in sound motion picture equipment were estab- 
lished in the Research Laboratories of RCA and affiliated companies 
largely upon the basis of reproduction of music. In the early research 
work, the necessity for eliminating flutter and other forms of distortion 
was realized from tests made by recording piano and orchestral music, 
the quality of which was judged by eminent musicians and others 
capable of detecting distortions of tone. Thus, there was an early, 
and there has been a continuous, appreciation by research and de- 
velopment engineers, of the necessity for producing apparatus that 
would avoid distortion whether arising from mechanical imperfec- 
tions such as sprocket flutter or from circuit limitations. It is un- 
fortunate for the progress of the industry that, with few exceptions, 
most studio technicians have been so absorbed in the production 
problems that they have probably not fully realized the seriousness of 
distortions present in most recording and reproducing systems. 

Through field organizations maintained throughout the world for 
servicing theater installations and assisting licensees in the operation of 
recording equipment, the necessary experience for guiding new de- 
velopments has been gained. That this experience has been ex- 
tensive is indicated by the fact that although only 300 RCA repro- 
ducer installations had been made by January, 1930, approximately 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RCA Manufacturing Co., Camden, N. J. 

103 



104 M. C. BATSEL fj. a M. p. E. 

5860 were in operation by January, 1935. One hundred and three 
recording channels have been supplied to RCA licensees. Continued 
research and experience with commercial equipment have proved that 
the refinements in equipment demanded by the original ideals are 
vitally necessary for the more satisfactory reproduction of sound, 
upon which depends the future progress of the sound picture. 

Research has been, and is, constantly directed toward achieving 
possible improvements through applying the new discoveries in the 
various branches of physics and engineering. The results of this re- 
search and investigation into all methods of recording, which have not 
been limited to the variable-width system, are today's recording and 
reproducing devices, which more nearly approximate the engineer's 
goal of reproduction of the original sound to arouse in the minds of 
the audience the illusion of being present at the scenes depicted upon 
the screen. 

Some developments that resulted in a reduction of distortion are : 

(1) Film-moving mechanisms for both recording and reproducing equipment 
that are free from objectionable speed variations. Elimination of flutter caused 
by film-gate construction and ripples produced by sprocket teeth. 

(2) Improvement of the light-modulating system used in photographic re- 
cording so that it is now adequate for a frequency range extending to 10,000 
cycles. 

(3) Improvement of amplifiers, through the development of new types of 
vacuum-tubes, improved transformers, and resistors. 

(4) New laboratory devices for analyzing the causes of distortion, so that 
they might be eliminated. 

(5) New types of microphones. These are of the velocity type, and have a 
smoother response over a wider frequency range than previous types, and fulfill 
the requirement for a directional microphone having characteristics independent 
of frequency. In these microphones not only have the immediately realizable 
improvements been considered; we have also considered some of the more 
fundamental factors essential for further improvement in sound recording. 

A further explanation of the value of the velocity or ribbon micro- 
phone in eliminating distortion is probably warranted. The percen- 
tage of reflected or reverberant sound to which a microphone responds 
in a room having uniformly reflecting boundaries is a function of the 
distance from the source and the solid angle from which sounds are 
received. The more limited the angle, the farther the microphone can 
be used from the source, for the same apparent distance as judged by 
listening. 

The directional characteristics of pressure-operated microphones 



Aug., 1935] RECORDING Music FOR PICTURES 105 

are such that sensitivity at low frequencies is uniform in all directions 
and limited to a rather small solid angle at high frequencies. This 
characteristic results in a different ratio of direct to reflected sound as 
the frequency changes. The effect is the same as though the micro- 
phone were much closer to the source at high frequencies, and pro- 
duces a distinct "hardening" of the sound. L. E. C. Hughes, lecturer 
in Electrical Communications, City and Guilds Engineering College, 
London, has used the term "acoustic distortion" to describe this 
effect. It is this effect that explains the objectionable sibilants that 
are largely due to the directional characteristics of microphones and 
loud speakers. Unless the sibilants become hard and objectionable, 
tests have shown that speech is more realistic and pleasing when the 
frequency range is extended. 

Velocity microphones, besides eliminating the contribution of the 
microphone to the hardening effect, possess other desirable character- 
istics. The directional characteristic can be utilized to differentiate 
between accompaniments and solo parts in musical recordings. Its 
use lends a uniform quality to the various instruments in an orchestra, 
usually located at different angles with respect to the microphone. 

Since the human ear is able to hear frequencies above 10,000 cycles 
per second, it seems logical that reproducing systems should be im- 
proved to include a similar range. An extension of the frequency 
range is practicable with the equipment developed, but without modi- 
fications of technic an extension of the frequency range may give re- 
sults not entirely satisfactory to the ear. 

Close observations have been made of the performance of theater 
reproducing equipment when reproducing from films recorded upon all 
types of commercial equipment now in use. These observations have 
led to certain conclusions with respect to the methods employed in the 
recording studio as well as to the relative merits of the various sys- 
tems used for recording. Mendoza, 1 in 1933, pointed out some of the 
defects in musical reproduction that were apparent to him at the 
time, such as loud speaker arrangement, acoustical characteristics of 
stages, sprocket hole modulation, limited frequency range, "fuzz," 
and "edge." With the equipment now available to recording studios, 
the defects to which he referred that could be attributed to the equip- 
ment have been to a great extent overcome. 

The lack of resonance or suitable reverberation has not been ade- 
quately considered. Theater reproducing equipment must be capable 
of reproducing the talking portions of a picture satisfactorily. By this 



106 M. C. BATSEL [J. S. M. P. E. 

is meant that intimate scenes, or scenes in which the observer is ap- 
parently very close to the actors, must be reproduced so that the 
auditors hear as though they were correspondingly close. This effect 
can be attained only when the reproducing system, including the 
auditorium, is free of resonances and appreciable reverberation. 
These necessary conditions can be fulfilled with non-directional speak- 
ers in small rooms acoustically treated with absorbent materials; or 
in large auditoriums having sound absorbing materials upon the rear 
walls, with directional loud speakers that direct the sound sufficiently 
to permit a high ratio of direct to reflected sound to reach the auditors. 

These conditions are not desirable for musical reproduction. The 
most desirable arrangement would appear to be to use two sound- 
tracks and two complete reproducing systems so that the dialog 
might be reproduced over a system similar to that now employed, 
and the music through a system utilizing an entirely different speaker 
arrangement, preferably an arrangement that would diffuse the sound 
and spread the sources over a greater area so as to make effective the 
reverberation of the auditorium in which the music is heard. For the 
present time, the best musical results can be attained by adequate 
reverberation and good tonal characteristics of recording stages. 

In support of the necessity for paying greater attention to the tone 
quality of musical recordings and the effect of the construction of 
scoring studios, use is made of information contained in Planning for 
Good Acoustics, by Bagenal and Wood, first published in London, in 
1931, a book that should be useful to all sound engineers. 

The history of music in relation to buildings shows that tone design has de- 
veloped not in the open air nor in the laboratory, but in the church, the opera 
house, and the concert hall. By their longer or shorter reverberations, those 
buildings have favored this or that type of music, but at all times they have set 
standards of tones which have been recognized. 

These standards should not be less recognized today in our efforts 
to reproduce music pleasing to the audiences. The famous music 
halls of the world, it appears, were not accidental in design, nor does 
it appear that reverberation alone was considered in the design. The 
authors of this book have designated tone to express "an energy 
condition as will enhance the characteristic of the main groups of 
musical instruments." This quality, it seems, is difficult to define, 
yet it is unmistakable to those who are sensitive to it. 

If there is present some wood paneling (or some other system in 
which the internal losses are not very great), it will vibrate whenever 



Aug., 1935] RECORDING Music FOR PICTURES 107 

a sound is produced having a frequency near its resonance frequency. 
The effects of these systems may be of considerable importance in 
recording studios. The resonances should be within the range to 
build up the fundamentals of the instruments. If the surfacing of the 
panel is smooth, overtones and harmonics will be reflected, and thus 
the brightness of the music be improved. 

The effect of the paneling is to increase the apparent loudness, improve the 
tone of the instruments, and to brighten the musical quality. 

It is, therefore, especially useful where the reverberation time is 
short. The authors of the book state that 

It is controversial whether the brightness secured in this way is a real substitute 
for the fullness of tone only available with adequate reverberation, but it is 
greatly superior to the deadness and dullness produced by the lack of both. 

It is probable that somewhat less than the optimal reverberation 
time may be fairly satisfactory when resonances of suitable wooden 
paneling are effective. This statement is based upon experience with 
a building that seems to have been famous for its acoustics. It is 
stated that Mozart, dementi, Schumann, Mendelssohn, and Wagner 
have conducted in the old Gewandhaus at Leipzig, and considered it 
to be very satisfactory. The reverberation time with full audience 
was only 1.5 seconds. The resonance of the ceiling and of the stressed 
wooden walls compensated for the short reverberation time. It 
seems that the large area of wood paneling was quite satisfactory for 
stringed instruments, but not as effective as reverberation for some of 
the other instruments, particularly the flute. 

Conditions in the scoring studios, however, can undoubtedly be 
greatly improved by applying some of the principles that have gov- 
erned the design of the satisfactory music halls of the world. The 
characteristically thin and unpleasant music usually heard in sound 
motion picture theaters could be improved during reproduction only 
by employing reverberation in the theaters. This is not practicable 
for talking pictures. 

The improved scoring stage acoustics would affect the performers 
equally as well as the auditors in the theaters. The performer hears 
the sound coming back to him as well as the sound from the instru- 
ment. This seems to be important, as it serves to guide the player in 
voicing the notes played or sung. A good room gives the performer a 
sense of power which he can control to produce the best effect. 



108 M. C. BATSEL 

Music is a sequence of tone relationships modified by the player and the room 
together. 

The microphone arrangement used for recording may influence 
greatly the liveliness of scoring stages without affecting reverberation 
time, resulting in a full and rounded effect that may be desirable 
under some conditions. A velocity microphone placed so as to be 
insensitive to direct sound will respond to the reflected sound, and 
thus increase the liveness. The effect can be conveniently controlled 
by the mixer. Recording can be done upon one film or through 
separate channels, and the mixing done in re-recording. 

REFERENCE 

1 MENDOZA, D.: 'Practical Problems in the Recording and Reproduction of 
Music for Motion Pictures," /. Soc. Mot. Piet. Eng., XX (Jan., 1933), No. 1, 
p. 79. 

DISCUSSION 

MR. LAMBERT: There are two different philosophies, I believe, in recording 
music for pictures. One is to create the illusion of reality in connection with the 
picture; the other is to record the music for its own sake. Suppose, for example, 
you wish an under-score for an intimate scene; if the scoring is reverberant, it 
tends to draw attention away from the scene that is being portrayed in dialog. 
The music, instead of adding to the scene, may detract from it. On the other 
hand, a very reverberant effect may be desired, as in the very effective long shots 
in the Madame Butterfly scene in One Night of Love. This quality was not as 
effective with the close-ups, however. We must keep in mind the use to which 
we are going to put the music when deciding how much reverberation is required. 

MR. BATSEL: Undoubtedly the perspective must be preserved. It would seem, 
from observations I have made, that a picture that is primarily musical had 
better avoid scenes that are too close. 



IMPROVEMENTS IN PLAYBACK DISK RECORDING* 
G. M. BEST** 



Summary. A method of recording sound upon cellulose acetate disks for play- 
back purposes in studio production is described. Means of adapting old wax record- 
ing equipment to the new recording material are explained, and comparisons with 
soft wax and shellac pressings are made. Location recording equipment for playback 
disks is made possible at minimum expense by the use of acetate disks. 

The revival of the screen musical several years ago brought with it 
an increased use of the system of disk playbacks whereby the music 
comprising the orchestral arrangements with singing or dancing 
could be pre-recorded and played back upon the set in synchronism 
with the camera, while the action was being photographed. It 
is not intended to discuss the relative merits of various uses of play- 
backs, but a brief description of the requirements for playbacks is 
given for those who are not familiar with them. 

To handle pre-recorded playbacks upon the set, a 33 Vs rpm. turn- 
table operated by a motor synchronized with the cameras is required, 
plus a reproducing system consisting of a reproducer, amplifiers, and 
loud speaker. The disk records are so marked that the needle can 
be placed in any predetermined groove a few seconds of playing time 
ahead of the part that is to be used, to avoid waste of film. When the 
turntable and cameras have reached the standard speed, the actors 
sing or speak their lines in unison with the sounds coming from the 
loud speaker, the sound mixer generally watching the lip movement 
closely to make sure the actors are in step with the pre-recording. 
Frequently, in order to aid in cutting the picture, a sound-track is 
recorded simultaneously with the playback, being a composite of the 
reproduced sounds from the playback, and the voices of the actors. 
This track is discarded after the picture has been edited. An addi- 
tional use of playbacks is to photograph and record one camera 
angle in a scene where several angles are needed, and from a soft wax, 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Warner Bros.-First National Studios, Burbank, Calif. 

109 



110 G. M. BEST [J. S. M. P. E. 

to play back the recording in synchronism with the camera while 
photographing other angles. 

Where the pre-recording could be done sufficiently in advance of 
the time the playbacks were required upon the set, it was customary to 
record the numbers in soft wax so they could be processed and shellac 
pressings furnished, or the sound-track was processed and played back 
through a reproducing dummy, generally from the recording build- 
ing rather than upon the stage. Use is also made in some studios of a 
toe-recorded negative which is available for playback within an hour 
after it is made, provided it is recorded in the studio. If the play- 
backs were required immediately, it was necessary to cut a number of 
soft waxes simultaneously, playing them back from the recording 
room. 

These playbacks were expensive because of the additional personnel 
involved, the short life of each wax, and the fact that the playback 
disk was not upon the set, thereby slowing up production. Immedi- 
ate playbacks upon location required at least a two-machine set-up, 
with a cumbersome and expensive truck and personnel. 

In view of these handicaps, the cellulose acetate disk was brought 
into use about a year ago, and at some studios it has completely sup- 
planted all other methods for obtaining playbacks. It has made pos- 
sible immediate playbacks on location at a small cost, has cut the 
cost of playback disks to a small fraction of the former total, and has 
saved many hours of production time. The acetate disk is not new, 
having been used for several years in other lines of recording such as 
air checking for radio stations, so that they are available commercially 
in several types, the most common consisting of an acetate coating 
upon both sides of a steel or aluminum disk which usually is furnished 
in a 12- inch diameter. For motion picture use, an aluminum disk 
coated with a black acetate has proved the most satisfactory, as it is in 
appearance much like the shellac pressings previously used and can 
be marked with white grease pencil so that any section of the record- 
ing can be played over and over again from the same start mark. 
The acetate coating is about 7 mils thick on each side, providing ade- 
quate material to cut a groove of standard width and depth. As the 
disks are double surfaced, the cost per disk is actually less than the 
cost of shaving two 12-inch waxes. 

Cutting acetate surfaces involved certain changes in recording tech- 
nic. For soft wax recordings, the Western Electric lateral-cut re- 
cording head had been in use since the early days of talking pictures, 



Aug., 1935] 



PLAYBACK DISK RECORDING 



111 



and was designed to cut soft wax at a cutting pressure of approxi- 
mately 20 grams, or slightly less than one ounce. The mechanical 
filter components of the recorder were so designed as to function with 
that cutting pressure, and trial recordings using this cutting head upon 
acetate verified the conclusion that the greatly increased cutting re- 
sistance of the acetate required so much additional cutting pressure 
as to upset the filter characteristics to such an extent as to lose the 
intelligibility of speech. 

Some idea of the change in frequency characteristic of this recorder 
when used on acetate can be gained by referring to the curves in 
Fig. 1. Curve 1 shows the characteristic of the recorder when cutting 
soft wax, the data for the curve having been obtained by playing back 




Frequency 

FIG. 1. Characteristics of recorder cutting wax and acetate 
disks: (1) soft wax, standard recorder; (2} soft wax, modified 
recorder; (5) acetate disk, standard recorder; (4) acetate disk, 
modified recorder. 

a soft wax frequency recording. Curve 3 shows the same recorder 
used to cut an acetate surface and played back with a 4- A Western 
Electric reproducer, using the same amplifier as for the soft wax play- 
backs. Curve 3 shows the damaging effect of the cutting resistance of 
the acetate at frequencies above 1000 cycles. 

Fortunately, one of the components in the mechanical filter in the 
recorder consists of a pair of balancing springs which serve to keep 
the armature in center. By tightening these springs, the character- 
istic of the recorder was tilted at the high end to such an extent as to 
give a very satisfactory playback from acetate. The extent of this 
tilt can be seen in curve 4, while curve 2 shows the effect that tighten- 
ing the springs had upon the soft wax characteristic. It is obvious 



112 



G. M. BEST 



[J. S. M. P. E. 



from inspection of both curves 1 and 3 that neither represents what 
we should regard today as a good frequency characteristic for theater 
reproduction; but for playback purposes in a production set-up, 
intelligibility is the prime factor, and a predominance of high fre- 
quencies is considered of advantage to the actors in following the 
dialog or music. Equally obvious is the fact that a recorder adjusted 
for acetate disks is unfit for soft wax. 

The cutting stylus for acetate recordings can be either the sapphire 
jewel used for soft wax, or a tungsten stylus shaped somewhat 
like the sapphire. The sapphire wears well provided there are no 
impurities in the acetate, especially abrasive coloring matter; but, 




FIG. 2. 



Direction of Surface Travel 
Angles of cutting stylus with respect to disk. 



unfortunately, the commercial acetate supply has occasional bits of 
foreign matter imbedded in the surface. Also, when the feed screw is 
stopped at the end of a take, if the stylus is not immediately lifted 
oft the disk, it will cut through the acetate down to the metallic base, 
and a sapphire jewel is quickly chipped if it comes into contact with 
the metal. A tungsten stylus will not chip on aluminum, and will 
cut several acetate disks before it needs resharpening. The cutting 
pressure required for acetate is approximately 85 grams and, due to 
the constant and uniform cutting resistance of the acetate, no advance 
ball is required, the depth of cut being governed by the balance 
weights upon the end of the recording instrument. The position of 
the stylus is somewhat different from that used for soft wax recording, 



Aug., 1935] PLAYBACK DISK RECORDING 113 

for the cutter must scoop the material from the acetate disk in order 
to avoid chattering. If the stylus cutting face is exactly 90 degrees 
with respect to the disk surface, a dragging action will ensue, and a 
high-frequency chatter will be superimposed upon the recording. 
By placing the cutting face of the stylus between 86 and 87 degrees, 
with respect to the disk surface, as shown in Fig. 2, a smooth cut can 
be obtained. 

The sliver of material from the disk can be drawn into the vacuum 
system in the same manner as for soft wax, except that the suction 
tube is mounted to the rear of the cutting stylus rather than directly 
in front of it. If no vacuum is available the sliver can be allowed to 
pile up upon the disk, as it has a tendency to be drawn in toward the 
center of the disk and does not interfere with the cutting stylus; 
this phenomenon is particularly useful for location recording, as it 
obviates the need of a vacuum system. If the cutting stylus becomes 
dull, the sliver is not uniform, and consists of small particles mixed 
with dust. Tests made with styli of various forms of steel, especially 
carboloy, proved that tungsten was superior, the steel causing a 
feather-edge upon the walls of the groove and a resultant high surface 
noise. Using the tungsten stylus, the surface noise is approximately 
that of the best grade of shellac pressings. A tungsten stylus indi- 
cates dullness most readily by increased surface noise when the disk 
is reproduced, and also by narrowing of the groove during recording. 
Styli can be sharpened in a few minutes' time by placing them in a jig 
specially made to hold the stylus in position while the edges are 
honed. 

In reproducing acetate disks, the standard shellac disk playback sys- 
tem can be used without change other than to reduce the weight of the 
reproducer at the needle point to about 2 1 / z ounces. This is done by 
means of a counterbalance weight upon the opposite end of the sup- 
porting arm, and with this weight, standard medium-tone steel needles 
can be used with a life of about 300 play ings for the disk. If the 
weight of the needle exceeds 3 ounces the acetate is quickly cut 
through. Cutting and reproducing speeds up to 78 rpm. are quite 
practicable with the acetate disk, although the life of the stylus is 
considerably reduced by the higher cutting speed. 

Location recording for immediate playback is made possible with 
the acetate disk, as a recording machine can be mounted in a small 
trailer truck, towed by the film recording truck to the location. Out- 
side of adjusting the position of the truck so that it is approximately 



114 G. M. BEST fj. s. M. P. E. 

level, an operation that is not critical, there is nothing in the operation 
that is involved or liable in any way to hold up production. Distant 
locations have been efficiently handled in this manner, sufficient re- 
cording heads with sharpened styli being taken along to meet require- 
ments. With immediate playbacks available upon locations, or upon 
the stages, a resultant speeding up of production, saving in processing 
costs, transportation, and material has made the acetate disk an 
important part of the industry. 

DISCUSSION 

MR. THAYER: What is the angle between the stylus and the disk? If the angle 
is 87 degrees, does that mean that the stylus leans three degrees forward, or back- 
ward? 

MR. BEST: The stylus leans backward slightly. The angle is 87 degrees look- 
ing from the rear of the recording stylus. The stylus has to scoop the material 
out slightly. 

If it is exactly at right angles, it oscillates at about 5000 cycles; and if it leans 
slightly forward, it has a tendency to jump out of the material. In other words, 
if the angle is 85 or 84 degrees, it seems difficult to control the depth of the cut. 
There is almost an exact setting required, which is usually determined by the 
recordist when he is setting up the instrument. 

He cuts an unmodulated groove and regulates the height of the instrument 
above the disk, and adjusts the angle very nicely. He listens for the 5000-cycle 
oscillation; and when it disappears, he knows that he has the right setting. We 
do not actually measure the angle every time we set up the instrument. The 
adjustment is almost a matter of instinct with the recordist. 

MR. J. CRABTREE: What is the range of frequency that you reproduce? 

MR. BEST: About 5000 cycles; not above that. Even at 5000 cycles, it is 
down about 8 db. 

MR. J. CRABTREE: The quality sounds about as good as we hear in the average 
theater today. 

MR. BEST: That is a compliment, because we have not tried to make the 
quality as good as it might be. So long as the actors upon the set can under- 
stand the recordings, particularly in short dialog sequences, that is sufficient. 
Unless the sibilants are heard the actors have difficulty in following the words that 
they spoke a few minutes before. 

MR. J. I. CRABTREE: It isn't quite clear as to how you utilize the playback. 
Can the actors synchronize their own lip movements? 

MR. BEST: Yes. We make playbacks of dialog as well as of singing and 
music. It is more or less standard practice in recording a musical to pre-record 
the number upon a scoring stage with orchestra and voice. The record you have 
just heard was supposed to simulate a dance band in a hotel, with the usual vocal- 
izing between the orchestral numbers. It was all recorded upon the scoring 
stage two days before the picture was shot. On the set where the picture was to 
be made, we installed a turntable operated by a motor synchronized with the 



Aug., 1935] PLAYBACK DISK RECORDING 115 

cameras, turned over by the same distributor, and the loud speaker was placed 
in such a position that the singer could hear it very plainly. 

In order to aid in the cutting, we have a microphone almost anywhere in the 
set, outside the picture, and record the playback coming from the horn as well as 
the actor's voice, to enable the cutter to cut the picture so that it will fit in ac- 
curately with the original sound-track, which was recorded at the same time the 
celluloid disk was made. Afterward, that track is thrown out. 

MR. J. I. CRABTREE: How far is the actor usually out of synchronism? 

MR. BEST: The sound mixer upon the set stands as close to the actor as he 
can, and watches the lip motions. Usually a representative of the music depart- 
ment is there also; a take will not be approved until both are satisfied that the 
synchronism is perfect. I have seen sequences as long as four minutes, in which 
a song with several verses and a chorus were sung, and at no time during the pro- 
jection of the picture afterward could one tell that it was made to a playback. 

PRESIDENT TASKER: I should like to repeat some thoughts of Mr. P. H. Evans 
of New York. 

Heretofore, photographic take playbacks have been made from soft wax, and 
hence the wax was normally located in a recording room remote from the stage. 
This required that when a rehearsal was about to begin a signal be transmitted 
to the recording room and the recording started. If a breakdown occurred 
in the rehearsal, a signal to stop went to the recording room. Then the wax had 
to be reset and restarted when the actors were ready. 

When first using this method, Mr. Evans found that the fact that the 
record could be very quickly made and brought upon the stage, and there placed 
upon a turntable in full sight of the actors, was of immense facility in speed- 
ing rehearsals. It almost entirely eliminated the chain from actor to director 
to assistant-director to recorder and vice versa, to take care of one of these starts 
and stops. 

Suppose that the artist was a dancer. He merely nodded to the boy standing 
beside the set to start the record at the approved place. If a breakdown occurred, 
a flick of the hand would stop it, or start it again, or pick it up at any spot. The 
facility of using the device for rehearsing playbacks was a startling advance in this 
class of activity. Who originated the method, I don't know. Mr. Best's paper, 
of course, is concerned mostly with improvements in recording methods. 

MR. BEST: It has been in effect, so far as I know, since the Fall of 1928. The 
first time I ever saw it used was in Warner Brothers' production of the first musical 
they ever made, The Desert Song. They used standard shellac playbacks, pro- 
cessed from soft wax, and they waited until the next day to shoot the picture, 
which required a twenty-four hour delay. We are now able to play the record 
back immediately. 

On location a few weeks ago in a down town Los Angeles theater, we mounted 
the location truck alongside the film-recording truck in the alley, and playback 
records were made while one camera angle was being photographed and the sound- 
track recorded. The disk was rushed inside and placed upon a turntable, and 
the other camera angles of the same scene were made to the playback, resulting 
in no delay whatsoever to the production set-up. 

With the former method, if immediate playbacks had been required, we should 



116 G. M. BEST 

have had to construct a soft wax recording set-up, with the bulky truck and the 
difficulties usually encountered in levelling the soft wax machine. 

The great advantage of the acetate record is that recordings can be made with 
the truck leaning at an angle ; so long as the recording instrument is rigidly held 
in its mounting. If the sliver of acetate piles up, the recording stylus plows its 
way through without jumping or in any way affecting the recording. So, although 
it is rather a haywire-looking device upon location, it works and saves a lot of 
time. 

Before we started using the method I can recall having the film made upon 
location, rushed to the laboratory, and the finished print dubbed to soft wax and 
then sent to the disk-processing plant; six hours later we received the shellac 
playbacks, which we rushed by motorcycle to the location. If the location was 
far away, it was the next day before the other camera angles could be shot upon 
the same set. 

MR. HANSEN: What is the volume range? 

MR. BEST : The surface noise determines the volume range ; the actual surface 
noise of this disk, as you could hear, was about the same as what would be ob- 
tained with a 6- or 8-db. noise reduction on the film. I should say 35 or 40 db. 
would be the maximum. The low surface noise of the acetate disk is of advantage 
on location because the same stylus can be used until it is very dull. The only 
result of using a dull stylus is to increase the surface noise, and with very little 
surface noise to start with, a considerable increase can be tolerated. 



A CONTINUOUS OPTICAL REDUCTION SOUND PRINTER* 
O. SANDVIK AND J. G. STREIFFERT** 



Summary. A continuous optical reduction sound printer is described, which 
prints by optical means from standard 35-mm. film to standard 16-mm. sound-film. 
Since the longitudinal reduction of the sound-track is greater than the lateral reduc- 
tion, an anamorphote optical system is required. In the present case this optical 
system consists of a combination of spherical and cylindrical lens elements. Data 
are given which show the change in the frequency characteristic of the 16-mm. sound 
print as a function of the 35-mm. negative shrinkage. 

The making of 16-mm. prints from 35-mm. picture negatives is 
now a relatively old and well-established practice. In the more re- 
cent problem of transferring sound from 35-mm. film to 16-mm. film, 
however, the problems are somewhat different and the requirements 
somewhat more stringent. 

Two general procedures for carrying out this operation suggest 
themselves, namely, by electrical re-recording from 35-mm. film to 
16-mm. film, and by reduction printing from 35-mm. film to 16-mm. 
film. 

The relative merits of the two methods have been a subject of 
much discussion. It is now generally agreed that the printing of the 
sound is the more practical procedure for making 16-mm. release 
prints. This can be accomplished either by reducing the sound di- 
rectly from 35-mm. film to the 16-mm. release print, or by reducing 
from 35-mm. film to a 16-mm. film from which release prints are made 
by contact. Making the 16-mm. release print directly from the 
35-mm. negative record should lead to somewhat better results. The 
relative costs of prints made by the two methods would depend 
somewhat upon the number of prints made and the equipment avail- 
able in any given laboratory. 

Descriptions of several types of continuous reduction printers 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. Communica- 
tion No. 554 from the Kodak Research Laboratories. 
** Eastman Kodak Co., Rochester, N. Y. 

117 



118 



O. SANDVIK AND J. G. STREIFFERT [J. S. M. P. E. 



have appeared recently in the literature. One type 1 consists basic- 
ally of a single shaft carrying 35-mm. and 16-mm. sprockets bearing 
the correct diametrical ratio, these sprockets serving to draw the re- 
spective films through stationary gates. Light from the illuminating 
lamp passes through the 35-mm. gate, and a suitable optical system 
forms an image of the 35-mm. sound-track upon the 16-mm. raw 
stock in its gate. 




FIG. 1. 



View showing the 35-mm. side of the reduc- 
tion printer. 



Another paper 2 has described an optical reduction printer which is 
a modification of a standard 35-mm. to 16-mm. re-recorder. The 
electrical reproducing and recording elements in this case have been 
replaced by appropriate optical and illuminating systems. A sepa- 
rate filtered drive is used for each film. 

The new Eastman Model B reduction printer is a modification of 
an earlier model which has given very satisfactory performance over 



Aug., 1935] 



OPTICAL REDUCTION SOUND PRINTER 



119 



a period of four years. The original design necessitated the use of a 
filtered flywheel drive upon the main shaft carry ing the printing sprock- 
ets, whereas the new model has been designed for use either with or 
without such a flywheel. Since the recent settlement of the flywheel 
patent litigation, the use or disuse of a flywheel is a question only of 
its technical advantages. Experience with the new model indicates, 
however, that the flywheel is unnecessary. 




FIG. 2. 



View showing the 16-mm. side of the reduc- 
tion printer. 



Figs. 1 and 2 illustrate the new printer from the 35-mm. and 16-mm. 
sides, respectively. It will be noted that there are three pairs of 
sprockets, serving the purpose of supply sprocket, printing or 
main sprocket, and take-up sprocket. These sprockets have the 
correct diametrical ratio to drive the films at the proper relative 
speeds. Each pair is mounted upon a common shaft, thus assuring 
uniform relative motion of the films at all points. 



120 



O. SANDVIK AND J. G. STREIFFERT [J. S. M. p. E. 



A single high-grade worm-reduction couples the constant-speed 
motor to the shaft carrying the main printing sprockets. A separate 
worm-reduction drives a vertical shaft which, in turn, drives the 
supply and take-up sprocket shafts through bevel gears. High-grade 
chains connect the take-up spindles with the take-up sprocket shaft. 
The precision with which the entire machine has been built has re- 
sulted in a very light and uniform load upon the over-size motor. 
This assures constancy of speed, uniformity of exposure, and a 
minimum of mechanical wear. 

Considerable difficulty has been experienced in obtaining sprockets, 

both 35-mm. and 16-mm., of 
sufficient accuracy and uni- 
formity in regard to pitch, shape 
of tooth, and quality of the tooth 
surface which comes into contact 
with the film. A new method of 
cutting sprockets is being de- 
veloped which, it is believed, 
will overcome this difficulty. 
Maximum life of sprockets and 
film is assured by the use of a 
minimum of film tension. 

The 35-mm. sprockets have 
been designed to match film 
having a shrinkage of 0.25 per 
cent and to accommodate a 
shrinkage range of about 1.75 per cent. Hence, they will accom- 
modate film having a maximum shrinkage of about 2.0 per cent with- 
out interference. The 16-mm. sprockets are cut to match film of 
zero shrinkage and have a range of about 0.7 per cent. The matter 
of shrinkage will be discussed more fully later. 

In order to eliminate any possibility of distortion or disturbance 
of the optical system due to vibration, the printer has been built 
especially rigid and substantial. This, combined with the use of a 
motor equipped with a floating power type of mounting, has resulted 
in a machine which is unusually free from vibration. 

Although 35-mm. and 16-mm. films run on opposite sides of the 
machine, both films can be threaded entirely from the front. Film 
speeds are 70 feet per minute for the 35-mm. film or 28 feet per minute 
for the 16-mm. film. 




FIG. 3. Schematic diagram of the 
illuminating system. 



Aug., 1935] OPTICAL REDUCTION SOUND PRINTER 121 

The illuminating system is shown diagrammatically in Fig. 3. 
It consists of a spherical lens, L b which images the lamp filament in 
a second spherical lens, Lz, Z/2 forming an image of the uniformly 
illuminated lens LI upon the 35-mm. film. The lamp used during 
tests was a standard 260-watt, 52-volt prefocus projection lamp. 




FIG. 4. Top-front view showing the location of the 
optical system with respect to the printing sprockets. 

Although not shown in the figure, a spherical mirror is used behind 
the lamp, and is located with its center of curvature in the plane of 
the filament. This increases the light very materially. A lamp of 
higher wattage, such as a biplanar lamp, can be used if for any reason 
it were found that more light was required. The present 260-watt 
lamp, however, gives more than ample light for any printing material 
now in general use. 



122 



O. SANDVIK AND J. G. STREIFFERT [J. S. M. P. E. 



Owing to the hub of the 35-mm. main sprocket, it is necessary to 
bring the light into the sprocket above the hub and deflect it by means 
of the prism P 2 so that the light is incident normally upon the film 
within the printing aperture. 

The printing aperture measures 101 mils wide and 65 mils high. 
This height of aperture accounts largely for the high exposure avail- 
able. At the same time, however, this height of aperture makes it 
essential that the longitudinal magnification of the imaging system 
be very accurately adjusted, since the magnification determines the 
speed at which the image formed upon the 16-mm. film travels. Un- 

less this image travel at the same 
speed as the 16-mm. film, there 
will be relative slippage between 
the image and the film during 
the period of exposure, with a 
consequent loss of definition. 
Both 35-mm. and 16-mm. films 
are guided at the printing point 
from the edge of the film nearer 
the sound-track. 

Fig. 4 is a close-up view of 
the imaging system, with covers 
removed. The components are 
shown diagrammatically in Fig. 5. 
Since the transverse magnifica- 
tion required is 0.845, whereas 
the longitudinal magnification is 
0.400, the image-forming lens 



Li AND 1-5 ARC CYLINDERS 
L 4 IS SPHERICAL. 



FIG. 5. Schematic diagram of the 
optical system. 



must be of the anamorphote type. This is accomplished by means 
of the cylindrical lenses, Z, 3 and L 5 , which have power along the 
lateral meridian only and which serve partially to neutralize the 
power of the spherical objective lens L\ in the lateral direction. 
Hence, the lateral magnification is determined by the combined 
power of L 3 , L 4 , and 1/5, whereas the longitudinal magnification is 
determined by the spherical element L 4 alone, which is a lens of high 
quality. 

The large roof prism serves to direct the light around to the 
1 6-mm. side and to invert the image in the longitudinal direction so 
that the image formed upon the 16-mm. film moves in the same 
direction as the film. 



Aug., 1935] 



OPTICAL REDUCTION SOUND PRINTER 



123 



EFFECT OF NEGATIVE SHRINKAGE IN SOUND REDUCTION PRINTING 



In order to investigate the effect of shrinkage of the negative film 
upon definition, reduction prints were made of a series of constant- 
frequency records. These negatives were identical except that thf 
various samples differed in shrinkage, covering a range from 0.1 
per cent, that is, 0.1 per cent above normal pitch, to +1.5 per cent, 
or 1.5 per cent below normal pitch. This represents a much 
greater range of shrinkage than that which will be encountered in 
present 35-mm. motion picture film. 

The modulation of these prints was then measured with the micro- 
densitometer, and the results are shown graphically in Fig. 6. 




NEGATIVE SHRINKAGE IN PERCENT 

FIG. 6. Microdensitometer readings showing the 
variations in the definition of the prints with variations 
in the shrinkage of the negative. 



The effective slit width was so small that the scanning loss can be 
neglected. It is seen that maximum modulation is attained for a 
negative pitch of 0.1864, corresponding to a shrinkage of 0.32 per 
cent. This is slightly in excess of the shrinkage for which the 
35-mm. sprockets are designed. 

It is of particular interest to note that in a print made from a nega- 
tive whose shrinkage was 1.5 per cent, the loss in modulation in the 
print resulting from the negative shrinkage was only 2 db. at 4000 
cps. and 5 db. at 8000 cps. The loss increases somewhat more 
rapidly as the shrinkage becomes less than that for which the sprocket 



124 



O. SANDVIK AND J. G. STREIFFERT [J. S. M. p. E. 



is designed, but this is unimportant because in practice the shrinkage 
will rarely fall below this value. Audition tests on prints made from 
the above negatives indicated that there was no noticeable increase 
of sprocket modulation as the negative shrinkage increased. Hence, 
it is seen that satisfactory reduction prints may be made from nega- 
tives covering a wide range of shrinkage. 

In this connection it is interesting to compare the relative effects 
of negative shrinkage upon definition in projection printing and con- 



RELATIVE PRINT MODULATION IN OB. 
1 1 , 
In o M o 




























































































































































/ 




~^ 


\ 




























/ 








\ 


\ 
























/ 












\ 






















/ 
















\ 


















/ 




















\ 














X 

















































































































































1.0 O.S 

NEGATIVE: SHRINKAGE IN PER CENT 

FIG. 7. Relation between negative shrinkage and 
print modulation in prints made with a standard 35-mm. 
contact printer. The shrinkage of the printing stock 
was zero. Modulation was measured on a frequency of 
9000 cps. 



tact printing. Accordingly, prints on raw stock of zero shrinkage 
were made with a 35-mm. contact printer from a series of constant- 
frequency negatives differing only in the amount of film shrinkage. 
The results are shown graphically in Fig. 7. It is seen that the modu- 
lation of the prints is a maximum for a negative shrinkage of about 
0.3 per cent and falls off rapidly for values of shrinkage greater or 
less than this value. 



EFFECT OF PRINTING FACTOR IN REDUCTION PRINTING 

If a sensitometric strip is printed both by contact and by projection, 
and the prints are given the same development, the projection print 



Aug., 1935] 



OPTICAL REDUCTION SOUND PRINTER 



125 



will be found to have a considerably higher gamma than the contact 
print. This is due to the specular nature of the light forming the 
image in projection printing as compared to the diffuse nature of the 
light in contact printing. 

For variable- width records this increase in gamma is in general 
advantageous because of the increase in sharpness of the photographic 
image along its boundary. In the case of variable-density records, 
however, the nature of the problem from the photographic standpoint 



77 



O.I 0.3 O.A 0.5 06 

DIFFUSE: PRINT DENSITV 



FIG. 8. Relation between harmonic distortion and 
unmodulated density in variable-density reduction 
prints: negative density, 0.62; negative gamma, 0.63; 
print gamma, 2.0. 



is somewhat different, being more intimately related to the exposure- 
transmission characteristic curve of the photographic material. 

Since the relation between the exposure of the negative and the 
transmission of the print is a function of the effective density of the 
negative in printing, it is evident that the sensitometric conditions of 
exposure and development which lead to the best wave -form in con- 
tact printing are not necessarily the best conditions for projection 
printing. To illustrate this, a series of reduction prints, that is, 
projection prints, covering a range of densities made from a constant- 
frequency, variable-density negative of unmodulated density of 0.62 
and gamma of 0.63, was developed under conditions which would 
obtain for good picture quality, say, gamma, equal to 2.0. It was 



126 O. SANDVIK AND J. G. STREIFFERT 

found upon analyzing these prints that the diffuse print density at 
which a minimum of harmonic distortion occurs is about 0.4, as shown 
in the curves of Fig. 8. For a contact print from the same negative 
and of identical development, the density is normally somewhat 
higher than this and hence, in general, the sensitometric conditions 
for projection printing are somewhat different from those for contact 
printing. 

REFERENCES 

1 VICTOR, A. F.: "Continuous Optical Reduction Printing," /. Soc. Mot. 
Pict. Eng., XXIII (Aug., 1934), No. 2, p. 96. 

2 DIMMICK, G. L., BATSEL, C. N., AND SACHTLEBEN, L. T.: "Optical Reduc- 
tion Sound Printing," J. Soc. Mot. Pict. Eng., XXIII (Aug., 1934), No. 2, p. 108. 



THE TECHNICOLOR PROCESS OF THREE-COLOR 
CINEMATOGRAPHY* 

J. A. BALL** 

Summary. The development of the Technicolor process is reviewed historically 
with particular reference to the guiding motives and technical objectives. This leads 
up to a brief description of the three-color camera and the three-color imbibition 
printing process. Following this is a discussion of the photographic principles 
involved in color photography particularly as they apply to the Technicolor three- 
color process. 

In the earliest days of the Technicolor development, we recog- 
nized that the ultimate goal of workers in the field of color cinema- 
tography must be a process that would add a full scale of color 
reproduction to the existing black-and-white product without sub- 
tracting from any of its desirable qualities, without imposing any com- 
plications upon theater projection conditions, and with a minimum 
of added burden in the cost of photography and in the cost of prints. 
These considerations seemed clearly to indicate a three-color sub- 
tractive printing process capable of ultimate low cost of manufacture. 

In those days, most other efforts to develop a subtractive printing 
process made use of double-coated positive stock, invented about 
1912 by Hernandez-Mejia. We found a number of objections to 
the use of this stock; particularly, to the spatial separation of the 
two components, to the susceptibility to scratching during processing 
and projection, but most of all, to the impediment imposed upon an 
ultimate three-color result. 

Surveying the field, we chose to work upon the multi-layer, or 
monopack process, and the imbibition process. In a monopack 
process the several components are in successive layers, all coated 
upon the same side of the film strip. In the imbibition process, the 
several components consist of images formed in water-soluble dyes 

* Presented at a meeting of the Technicians Branch of the Academy of Motion 
Picture Arts and Sciences, Hollywood, Calif., May 21, 1935. 

** Vice-President and Technical Director, Technicolor Motion Picture Corp., 
Hollywood, Calif. 

127 



128 J. A. BALL [J. S. M. P. E. 

printed on, or rather into, a gelatin-coated film strip, much as 
colored ink images are printed upon paper in the process of photo- 
lithography. A multi-layer, or monopack, process can theoretically 
be used as a taking process and as a printing process; whereas 
imbibition, being a photomechanical process, is limited to use as a 
printing process and requires to be supplemented by a taking method, 
preferably one providing distinct separation negatives. As printing 
processes, both monopack and imbibition yield a final product con- 
taining all components upon one side of the film strip and with no 
limitation as to their number. Some fundamental and far-reaching 
work upon the monopack process by the late Dr. Troland, who at 
the time of his death was research director of Technicolor, resulted 
in the issuance in 1932 of Reissue Patent No. 18,680, containing two 
hundred and thirty-nine claims, broadly covering this field both for 
taking and printing. The imbibition process seemed to present a 
less formidable array of processing problems than did the mono- 
pack process, so we pushed its development with even greater effort. 

We found it necessary to split the problem into two stages. As the 
first step in an imbibition process it is necessary to prepare a film 
bearing images consisting of a raised relief of hardened gelatin. This 
relief image, or matrix, serves the same purpose as the etched copper 
or zinc plate of photolithography. First, we had to find out how to 
make a gelatin relief suitable for use as a printing plate. We decided 
to content ourselves temporarily with two components and to stop 
short of actual imbibition by making use of an intermediate process 
wherein two gelatin reliefs, produced upon thin celluloid, were glued 
together back to back and dyed in complementary colors. Prints 
of the Technicolor sequence in The Ten Commandments, and of 
Douglas Fairbanks' all-color picture, The Black Pirate, were made 
in this manner. 

Then, after having learned how to make gelatin relief matrices 
of good quality, we tackled the problem of making adequate trans- 
fers from those matrices. We had to learn how to prepare the blank 
film so as to permit imbibition without diffusion. We had to devise 
a transfer machine capable of handling film in long lengths and in 
quantities, and in which blank and matrix could be brought into 
registered contact and held there for several minutes while the dyes 
transferred. 

Simultaneously with work upon these various subtractive printing 
processes, we devised a camera that gave two-color separation nega- 



Aug., 1935] THREE-COLOR CINEMATOGRAPHY 129 

live images free not only from fringing and parallax but also 
from the harmful effects of celluloid shrinkage. In this camera the 
two images were in symmetrical pairs, one being the mirror image 
of the other. These were arranged upon a single strip of negative 
stock with both members of the symmetrical pair positioned ac- 
curately with respect to symmetrically adjacent pairs of perforations. 
The perfect geometrical symmetry of this arrangement is shrinkage- 
proof during the entire life of the negative. The very compact 
prism system of this camera permitted the use of relatively short 
focal length lenses. The aberrations of the glass path were taken 
into account in the computations for these lenses. 

Two-color imbibition prints were brought out commercially in 
1928, just about the time that sound swept the industry. We were 
then immediately faced with the necessity of combining color with 
sound. The only procedure obvious at that time was to make the 
sound-track identical with one or both of the picture components; 
but this would give a sound-track in dye, which would have varying 
absorption throughout the range of wavelengths to which photo- 
electric cells are sensitive. The response from such a track would 
then, of course, differ for one type of cell from that for another type 
and especially so in the case of a variable-density track. We 
avoided this problem by starting, not with a blank film, but with a 
strip of positive stock upon which the sound-track could be printed 
and developed in silver while leaving the picture area blank. Imbi- 
bition transfer of the picture components into this blank area could 
then take place. This method is capable of giving a sound-track 
absolutely identical to that used in the black-and-white art. Better 
yet, because of the complete separation of the sound-track technic 
from the picture technic, the necessity of any compromise between 
sound and picture quality is eliminated and ideal sound-track proc- 
essing conditions are possible. Many millions of feet of two-color 
imbibition prints with a silver sound-track were produced by Tech- 
nicolor in 1929 and subsequent years. 

We were now ready to move on to a three-color process. Since 
we had planned to do so from the beginning, we encountered no 
fundamental impediment in our printing process. Mechanically, 
we had merely to combine the imbibition paths in groups of three 
instead of in pairs. 

The proper choice of dyes presented more of a problem. In a 
two-color process many colors are compromised, so to speak, and 



130 



J. A. BALL 



[J. S. M. P. E. 



there is considerable choice as to the manner and extent of compro- 
mise. In a three-color process, the accuracy of reproduction is 
greatly increased and the freedom of choice is greatly restricted. 

An adequate three-color camera was an exceedingly difficult 
problem. Three-component taking methods that use only a single 
aperture (monopack, screen-plates, and lenticulated films) have 
advantage of economy of light and of mechanism, but they all 
have other disadvantages, particularly as regards separating or 
differentiating between the various components; and some of them 
present difficult raw-stock manufacturing problems. 



BIPACMTWO FILMS WIT 
EMULSION SURFACES IN 
CONTACT) FRONT FILM 

SENSITIVE TO BLUE 
AND CARRYING A SURFACE 
COATING OF A BLUE ABSORBING 
DYE, REAR FILM PANCHROMATIC 
RECEIVING RED IMAGE 



LENS 



SPUTTERED GOLD MIRROR CBETW/E EN PRISMS) 
DIVIDES LIGHT BETWEEN THE Two APERTURES. 




PANCHROMATIC FILM RLCEIYING 

GREEN IMAGE. 



MAGENTA FILTER TRANSMITTING 
RED AND BLUE LIGHT. 



GREEN FILTER TRANSMITTING 
GREEN LIGHT ONLY. 



FIG. 1. Arrangement of optical system and films in the three-color camera. 

On the other hand, cameras that split the light to three separate 
apertures, while photographically and optically simple, have the 
disadvantage of loss of light in the splitting process, long or compli- 
cated optical paths, increased size, and mechanical complexity. We 
chose as a favorable middle-ground solution an intermediate line of 
attack wherein three records are obtained at two apertures. 

Fig. 1 shows schematically the arrangement of the optical parts 
and films in this camera. In making use of a bipack at one aperture, 
we have incorporated means for the practical elimination of halation 
and also for the elimination of any dependence upon the surface 
coating of one of the films for the exact determination of our red 



Aug., 1935] THREE-COLOR CINEMATOGRAPHY 131 

light filter. Thus, two of the most serious faults of ordinary bipacks 
have been removed. 

To insure that there shall be no differential shrinkage among the 
three strips of negative, we specify that the celluloid base shall be 
of the low-shrinkage type, as made by the Eastman Kodak Company. 
This low-shrinkage celluloid base is of such quality that after process- 
ing the negative, including the manufacture of a volume of release 
prints, the shrinkage is approximately 1 / 8 of 1 per cent, with differ- 
ences in shrinkage among the members of a group of about x / 8 of 
the total shrinkage. This amounts to a small fraction of Viooo of 
an inch across the longest dimension of the picture and is therefore 
entirely negligible. 

A group of five lenses ranging in focal length from 35 mm. to 140 
mm. have been designed for this camera to our specifications by 
Messrs. Taylor, Taylor, and Hobson. The chromatic correction 
of these lenses has been designed to give, in cooperation with our 
film arrangement, three images of unusually high correction, thus 
compensating for the loss of definition in the red record of the bipack. 
The most notable feature of these lenses, however, is the inclusion 
in the 35-mm. design of what might be called the inverse telephoto 
principle, whereby the back focal length is considerably longer than 
the equivalent focal length. 

However, it is not the purpose of this paper to go into further detail 
as to the design and construction of the camera, but to move on to a 
discussion of the methods of operating the camera. First, however, 
a brief outline of the complete process as we now work it is perhaps 
desirable. 

The Technicolor three-color camera photographs the three primary 
aspects of a scene (red, green, and blue) upon three separate film strips, 
simultaneously, at normal speed, without fringe or parallax, in 
balance, and in proper register with each other. These separate 
strips are developed to negatives of equal contrast and must always 
be considered and handled as a group. 

From these color-separation negatives, we print by projection 
through the celluloid upon a specially prepared stock, which is then 
developed and processed in such a manner as to produce positive 
relief images in hardened gelatin. These three hardened gelatin 
reliefs are then used as printing matrices which absorb dye. This 
dye is then transferred by imbibition printing to another film strip 
which, when it has received all three transfers, becomes the final 



132 



J. A. BALL 



[J. S. M. P. E. 



completed print ready for projection. To carry on the process of 
imbibition, it is necessary merely to press the matrix film into close 
contact with a properly prepared blank film and hold it there for 
several minutes. Matrices, of course, can be used over and over 
again. 

The colors of dyes used in the transfer process must be the sub- 
tractive primaries, namely, minus-red (or cyan), minus-green (or 
magenta), and minus-blue (or yellow). The relation of the taking 
colors to the printing colors is made clear in Fig. 2. 



GREEN 



MINUS BLUE 
=RED , GREEN 

^YELLOW 



RED 




MINUS RED 
= BLUE &( GREEN 



MINU5 GREEN 
'RED $ BLUE 

^MAGENTA 



d 



BLUE 



Note: Solid lines join the Ph'mary 
Colors. ~ Dotted lines join the 
Complementary 5ubtracbVe Colors 



FIG. 2. Diagram showing the relation of the taking colors to the printing 

colors. 



(To show the manner in which the final print is built up, a short demonstration 
reel was projected. First was shown the sound-track and the yellow dye component, 
next the cyan component, then the magenta component, and finally the complete image.} 

The process just described is designed to reproduce whatever is 
placed in front of the camera, not only as to color but also as to light 
and shade. But even the best of reproduction procedures, even 
that of oil painting upon canvas, is rather severely limited as 
regards reproducing light and shade. The contrast from whitest 
white to blackest black in a painting is perhaps 1 to 32. Upon 
projection from transparencies, as in motion picture work, the range 
may be slightly greater, about 1 to 64, but in no case is the great 
range of sensitivity of the eye adequately reproduced. The art of 



Aug., 1935] THREE-COLOR CINEMATOGRAPHY 133 

painting and the art of photography then, have this in common: 
that they seek to suggest a great range of visual contrasts by a skillful 
use of the more limited contrasts available in the method of repro- 
duction. 

In color photography, all very full exposures tend to bleach out 
to white, and all low exposures tend to drop into black. A highlight 
upon a face in black-and-white photography can, in the final print, 
be merely the bare celluloid, and the result will be still entirely 
satisfactory; but if, in a color print, such a condition exist, the 
delicate flesh tint will, in that area, be bleached out to white, and the 
face will look blotchy. All areas of the face should, therefore, be 
reproduced in such a manner as to yield a good flesh tint. Very 
light make-ups, and oily make-ups having considerable shine, are 
apt to be troublesome. In any case, it is necessary to control the 
light and lighting contrasts accurately and to avoid "hot spots." 

The art of the color cinematographer is intermediate between 
that of the painter and that of the stage artist. The painter has 
to work with pigments having a limited range of contrast but has 
great freedom of choice as to composition. The stage artist works 
with light, and so does not encounter the pigment limitation; but 
he must select his costumes, backgrounds, etc., to be harmonious in 
a great variety of arrangements, most of which are more or less out 
of his control. In color cinematography the difficulties of both are 
combined; there is the pigment limitation combined with the com- 
parative lack of control of composition. To illustrate this difference 
let us take, for example, a scene wherein a figure clad in white is to 
be illuminated by red light, as from a fire which is not visible to the 
audience. The stage artist, in arranging such an effect, must have 
a suitable background for the figure when it is viewed from a great 
many different angles. In arranging his lights, however, he can call 
for more and more intense beams of red light until he has achieved 
the desired effect. If a painter is endeavoring to get the same effect 
in a painting, he can select a favorable pictorial composition, but 
to depict the red illumination he can use only the brightest red 
pigment in his palette. If he is dissatisfied with his first effort, he 
can not heap on more and more of his red pigment. Obviously 
nothing is to be gained in that manner. He can only improve his 
result by suppression of, or contrast with, the background. Now in 
color cinematography, the brightest red that is available is the full 
value of red pigmentation in the film, and this is obtained by full 



134 J. A. BALL [J. S. M. P. E. 

value of the magenta and yellow dyes without any cyan dye. These 
conditions result from full exposure of the red negative with no 
exposure in the green and blue negatives. If the color cinematog- 
rapher is not satisfied with this full pigmentation and endeavors to 
get a more intense red by piling on more red light in front of the 
camera, he merely over-exposes the red negative and begins to get 
some exposure in the green and blue negatives. The corresponding 
areas in the print tend to bleach out to white. The significance of 
the pigment limitation can be summed up in a very few words: if 
the desired effect can be shown in a painting, it can be photographed, 
and if it can not be painted, it probably can not be photographed. 
While no such brief statement is ever strictly true, this one contains 
such a large percentage of truth that it is worthy of being set up as a 
guiding principle. 

In color photography, it is necessary to operate at rather high 
levels of illumination. If one is not careful, this may lead to a condi- 
tion like this: given only relatively weak light-sources, one finds it 
necessary to use a great many of these sources, in order to attain an 
adequate level. The widespread distribution of these units then 
tends to kill all shadows and eliminate modeling on faces. If, then, 
the attempt is made to provide modeling by superimposing a localized 
shaft of light, as from a spot-light, the face is burned up, blotchy, 
and generally unrecognizable. The way out of this dilemma is to 
recognize that modeling should properly be produced by shadows, 
and to use fewer and brighter sources or to mass the sources of illumi- 
nation so that shadows have a chance to exist. In other words, it 
is just as important for the cameraman to determine directions from 
which light shall not come as it is to determine directions from which 
light shall come. 

While color contrasts will occasionally produce a pleasing result 
when flatly lighted, that is not the way to get sharp photography, 
nor in general, the most pleasing photography. The Technicolor 
process is capable of reproducing a full scale of contrasts and those 
effects of light and shade (chiaroscuro), and those directional effects 
so striking in black-and-white are even more effective in color. These 
considerations apply not only to the lighting of figures and faces 
but also to the design and lighting of sets. In the design and paint- 
ing of sets, the art director should have in mind the cameraman's 
problem of achieving the necessary light levels with a minimum 
number of sources of illumination. Under these conditions, it is 



Aug., 1935] THREE-COLOR CINEMATOGRAPHY 135 

always much easier to keep parts of a set in low key by keeping 
light away from them, than it is to paint them dark and then be 
forced to illuminate them strongly. 

This need for fewer and brighter sources is one of the reasons why 
we choose carbon arcs in preference to incandescent tungsten lamps. 
Another reason is the fact that only in the white-flame carbon arc 
and in sunlight do we find the correct balance of blue and red com- 
ponents for the photographic emulsions with which we have to work. 
If tungsten lamps were to be used, it would be necessary to throw 
away the excess red light by the use of blue glass bulbs or over-all 
niters. An additional reason for the use of arcs is that at the high 
levels of illumination which we require, the heat rays emitted by 
incandescent lamps are a serious problem. Arcs radiate more light 
and very much less heat. If incandescent units were properly 
filtered to correct the color of the light and to absorb heat rays they 
would undoubtedly be useful on special occasions. 

Special arc units have been developed by the National Carbon 
Company and Mole-Richardson, Inc., for use in connection with the 
Technicolor three-component process. They have been designed 
to solve some of the earlier difficulties with arcs, especially noise and 
flicker. The older types of arc also gave off some smoke which 
appeared as carbon dust in the air, but it is possible to incorporate 
absorptive means in the vents to absorb this smoke. The only 
drawback to the use of arcs is the necessity for "time out" for re- 
trimming, but this can usually be made to coincide with other "time 
out" activities, particularly if the head electrician works closely 
with the director. 

There is no danger of Kleig eyes when using arcs, provided only 
that a sheet of ordinary glass is between each arc and the eyes of 
the people. This is a simple enough requirement and entirely 
eliminates any danger. 

The required level of illumination is not very different from that 
which was in use by many black-and-white cameramen before the 
introduction of supersensitive film. We have devised methods of 
measurement of illumination levels for the guidance of the camera- 
man. 

Exterior photography divides itself into four classifications: 

04) Sunlight shots wherein the scenery is of maximum importance. These 
occur abundantly in travelogues and scenics and quite frequently in 
dramatic photography, especially in establishing long shots. 



136 J. A. BALL [J. S. M. P. E. 

(B) Sunlight shots wherein faces are of greatest importance. 

(C) Imitation sunlight exteriors built upon a dark stage and artificially 
illuminated. 

(Z>) Night exteriors. 

In group A there are pronounced differences between color photog- 
raphy and black-and-white photography because color photog- 
raphy can reproduce those pleasing color contrasts of sky, water, 
blue haze, foilage, beach, etc., which are almost entirely lost in 
black-and-white. Furthermore, there is always a strong directional 
effect to the sunlight with very pronounced shadows. A front 
cross-light is best in color, whereas a side- or back-cross would gener- 
ally be preferred in black-and-white. 

In class B it must be realized that few faces will stand the harsh 
lighting of the direct sun as in a front cross-lighted setting. So 
gauzes, diffusers, reflectors, and sometimes "booster" light, must 
be called into use. Conditions are then most favorable if the sun- 
light comes from behind the figure. This is true in color or in 
black-and-white. The skillful cameraman takes advantage of the 
changing directions of sunlight throughout the day to schedule his 
shots and angles for best results. Cooperation between director 
and cameraman in such cases is even more important than in the case 
of interiors. 

It is, of course, perfectly obvious that if artificial light is to be 
mixed with daylight, as in the case of "booster" light, the color 
of the "booster" light must approximate sunlight. Here again the 
use of carbon arcs in preference to incandescent lights is clearly 
indicated. One might wonder if the change in sunlight quality 
from morning to late afternoon might not show upon the screen in 
abrupt changes in color of successive scenes. We have found it 
generally possible to correct for such differences in the printing. 
Such correction, however, is not possible where one encounters 
simultaneously very yellow light from the sun with blue shadows 
illuminated from a clear sky. Such an effect will, of course, carry 
through to the screen, and a very beautiful effect it is. 

The set-ups of group C are very troublesome if the illusion of 
reality is of importance. This illusion almost always is important 
in a motion picture so that the artificialities of the usual stage lighting 
are scarcely acceptable at all. Shadows can perhaps still be painted 
upon buildings, walls, and backgrounds but of course not upon people. 
Nor can the shade of a tree be so imitated. What is really needed is 



Aug , 1935] THREE-COLOR CINEMATOGRAPHY 137 

a light-source of greater power than any now available. Pending 
the development of such a source, the sun promises to return to 
its former importance. In other words, sizeable sunlighted exteriors 
to be photographed in color had best be real. The difficulties of 
imitating grass, shrubs, etc., also argue in the same direction. 

In the case of night exteriors (class D), color has one great ad- 
vantage over black-and-white in that it is possible to contrast 
moonlight and lamplight, for example, by the use of blue and amber 
niters. 

Technicolor adds practically no complications to sound recording 
other than a somewhat noisy camera and the necessity of eliminating 
"whistle" from the arcs. If the camera is adequately blimped, the 
problem of camera noise is solved forthwith. The whistle caused 
by high-frequency ripples in the electric current coming from the 
commutators of direct-current generators can be practically removed 
by the combination of an alternating-current filter at the generator 
and additional choke-coils at the individual arc units. 

When we come to the trick department, however, color has its 
special problems. Fades, lap-dissolves, wipe-offs, etc., can all be 
made by duping all three negatives and taking pains to preserve the 
register, exposure, and contrast balance. Those methods of com- 
posite photography that depend upon color differences can not be 
used in Technicolor. The projection background process is, of 
course, ideal for trick shots in color. However, there is the problem 
of adequate illumination of the projection screen. So far, projected 
backgrounds have been used in Technicolor only in relatively small 
areas, such as through the rear window of a taxi or limousine. Even- 
tually, we hope to be able to work out means for handling projection 
backgrounds in very much larger sizes, but at present we are rather 
restricted. 

There is a general appreciation of the fact that "color is coming." 
When sound swept the industry several years ago, it meant the 
introduction of a new and different technic, and of men of new and 
different training. The sound engineer was the "big shot." The 
cameraman was locked in a padded cell with his camera, and the art 
director was told how he could and could not construct his sets to 
meet the new acoustic considerations. Conditions will be much 
more enjoyable for everyone concerned when color sweeps the in- 
dustry. The sound men will not be affected in any way at all, but 
the cameraman and the art director will be given new tools to work 



138 J. A. BALL 

with, whereby the value and importance of what they can contribute 
to a picture will be greatly increased. For these reasons it is to be 
expected that the technicians generally will be enthusiastic and 
cooperative with the rising tide of color. 

It is the policy of the Technicolor Company to organize and main- 
tain a nucleus camera department and color art department for the 
purpose of accumulating experience and disseminating information 
and advice as to the skillful and effective use of Technicolor. Be- 
yond this nucleus the policy is to invite cooperation from the studio 
organizations and especially from those cameramen and art directors 
who desire to continue to lead in their respective fields. These 
men will generally be surprised, first, at the extent to which their 
conscious sense of color has become atrophied through lack of use 
while working in black-and-white; second, at the speed with which 
they can regain it; and, third, at the utter inadequacy of black-and- 
white photography in comparison with good color photography. 

When our color was of inferior quality, we used to hear the ex- 
pression "color interferes with the drama." Since the introduction 
of the three-component process, the expression has been rapidly 
fading out of use. Good color assists good drama. Dr. Herbert 
T. Kalmus, President of Technicolor, has supported a liberal policy 
of research and development work since the organization of the 
company. This policy is continuing, and the work involves nearly 
all departments. We propose to continue to improve our product 
until the last doubter is swept off his feet. 






COLOR CONSCIOUSNESS* 

NATALIE M. KALMUS** 

Summary. Color constitutes another step in the steady advancement of the motion 
picture toward realism, the same principles of color, tone, and composition applying 
to the motion picture as to the art of painting. In order fully to appreciate the color 
picture, a "color consciousness" must be adopted, the lack of which is tantamount in a 
degree to color blindness. 

Monotony is the enemy of interest, a fact that argues for the color picture; but a 
superabundance of color is unnatural. Psychologically, colors fall into the "warm" 
and "cool" groups, and each color and shade has its psychological implications: red 
danger, blood, life, heat; green nature, outdoors, freedom, freshness; etc. To build 
up personalities and to harmonize emotions and situations, these principles must 
apply, even to the extent of "color juxtaposition," or the psychological relation of the 
various colors to each other. For example, of two adjacent or contiguous colors, each 
tends to "throw" the other toward its complement, considerably affecting the emphasis 
or import of the color. 

On the walls of the cave in Altamira in Spain are found paintings, 
boldly sketched in three colors by Paleolithic man some fifty thousand 
years ago. These prehistoric paintings are quite artfully executed, 
and show that the artist possessed a fine sense of color and a desire 
to indicate motion as well as form. Various animals are depicted 
with the use of a red clay, an ochre earth, and a black pigment. 
One picture shows a wild boar in a standing position. In another 
picture, nearby, to show the same animal in a gallop, two sets of 
legs have been used. This ingenious method of showing action 
indicates the inherent desire of the artist to show motion in color. 
This ambition has come down through the intervening years to the 
present day. Now we see the culmination of that idea motion 
pictures in color. 

From a technical standpoint, motion pictures have been steadily 
tending toward more complete realism. In the early days, pictures 
were a mere mechanical process of imprinting light upon film and 
projecting that result upon a screen. Then came the perfection of 

* Presented at a meeting of the Technicians Branch of the Academy of Motion 
Picture Arts and Sciences, Hollywood, Calif., May 21, 1935. 

** Color Director, Technicolor Motion Picture Corp., Hollywood, Calif. 

139 



140 NATALIE M. KALMUS [J. S. M. P. E. 

detail more accurate sets and costumes more perfect photography. 
The advent of sound brought increased realism through the auditory 
sense. The last step color, with the addition of the chromatic 
sensations, completed the process. Now motion pictures are able 
to duplicate faithfully all the auditory and visual sensations. 

This enhanced realism enables us to portray life and nature as it 
really is, and in this respect we have made definite strides forward. 
A motion picture, however, will be merely an accurate record of 
certain events unless we guide this realism into the realms of art. 
To accomplish this it becomes necessary to augment the mechanical 
processes with the inspirational work of the artist. It is not enough 
that we put a perfect record upon the screen. That record must be 
molded according to the basic principles of art. 

The principles of color, tone, and composition make painting a 
fine art. The same principles will make a colored motion picture a 
work of art. The precision and detail of Holbein and Bougereau, 
the light effects of Rembrandt, the atmosphere and arrangements of 
Goya, the color of Velasquez, the brilliant sunlight of Sorolla, the 
mysterious shadows of Innes all these artistic qualities can even- 
tually be incorporated into motion pictures through the medium of 
color. The design and colors of sets, costumes, drapes, and furnish- 
ings must be planned and selected just as an artist would choose the 
colors from his pallette and apply them to the proper portions of 
his painting. 

In order to apply the laws of art properly in relation to color, 
we must first develop a color sense in other words, we must become 
"color conscious." We must study color harmony, the appro- 
priateness of color to certain situations, the appeal of color to the 
emotions. Above all, we must take more interest in the colorful 
beauties that lie about us the iridescent brilliance of the butterfly's 
wing, the subtle tones of a field of grain, the violet shadows of the 
desert, the sunset's reflection in the ocean. By such observation 
and study we develop a sense of color appreciation and train our eye 
to notice an infinite variety of hues. 

Serious cases of color blindness are comparatively rare; yet, 
because the average person is not trained in color appreciation, a 
decided lack of color consciousness is not at all uncommon. In order 
to appreciate operatic or classical music, people study music appre- 
ciation. Color appreciation, as a study, is almost entirely neglected, 
although color plays a most important and continuous part in our 



Aug., 1935] COLOR CONSCIOUSNESS 141 

lives. The average person listens to music for only a short portion 
of the time, but every moment of the day he looks upon some form 
of color. 

In the study of color appreciation we have two classes of objects. 
On the one hand, we have Nature, with its flowers, skies, trees, etc.; 
on the other hand, we have man-made objects of all kinds, including 
art pictures. In the first class the color is already created, and it 
remains for us only to enjoy and appreciate. In the second class 
we can exercise a certain amount of selectivity. Because of the 
general lack of color knowledge, that selectivity is not always temp- 
ered with wisdom. If the color schemes of natural objects were 
used as guides, less flagrant mistakes in color would occur. The 
use of black and white, however, to the complete exclusion of all 
color, is decidedly not in keeping with Nature's rules. 

Natural colors and lights do not tax the eye nearly as much as 
man-made colors and artificial lights. Even when Nature indulges 
in a riot of beautiful colors, there are subtle harmonies which justify 
those colors. These harmonies are often overlooked by the casual 
observer. The most brilliant flower has leaves and stem of just the 
right hue to accompany or complement its gay color. 

As we grow in the understanding of color and its uses, we find that 
our color appreciation develops simultaneously. All the better 
things in life require a color consciousness for their fullest appreciation 
and enjoyment. 

The eye is the organ of perception. The impulses of light re- 
ceived by the retina are transferred over the optic nerve path to 
the brain, and we become conscious of light and dark, motion, form, 
and color. Vision is a sense of ancient lineage and of early develop- 
ment in the individual life. Its characteristic is the clearness and 
precision of the data it furnishes the mind. Compared to sight, 
the other senses are dull and groping. It is the sense by which we 
receive the greatest number of stimuli from the world about us. It 
is the sense which most frequently affects the nervous system, 
dominates the attention, and stimulates the mind. 

It is a psychological fact that the nervous system experiences a 
shock when it is forced to adapt itself to any degree of unnaturalness 
in the reception of external stimuli. The auditory sense would be 
unpleasantly affected by hearing an actor upon the screen speak his 
lines in a monotone. The mind would strive to supply the missing 
inflections. The same is true, but to a greater degree, of the visual 



142 NATALIE M. KALMUS [J. S. M. P. E. 

sense. A super-abundance of color is unnatural, and has a most 
unpleasant effect not only upon the eye itself, but upon the mind as 
well. On the other hand, the complete absence of color is unnatural. 
The mind strives to supply the missing chromatic sensations, just as 
it seeks to add the missing inflections to the actor's voice. The 
monotony of black, gray, and white in comparison with color is an 
acknowledged fact. It is almost a psychological axiom that monot- 
ony is the enemy of interest. In other words, that which is monoto- 
nous will not hold our attention as well as that which shows more 
variety. Obviously, it is important that the eye be not assailed 
with glaring color combinations, nor by the indiscriminate use of 
black and white. Again taking our cue from Nature, we find that 
colors and neutrals augment each other. The judicious use of 
neutrals proves an excellent foil for color, and lends power and 
interest to the touches of color in a scene. The presence of neutrals 
in our composition adds interest, variety, and charm to our colors. 
On the other hand, the presence of color in our picture gives added 
force to the neutrals, emphasizing the severity of black, the gloomi- 
ness of gray, the purity of white. 

From a broader point of view, the psychology of color is of immense 
value to a director. His prime motive is to direct and control the 
thoughts and emotions of his audience. The director strives to 
indicate a fuller significance than is specifically shown by the action 
and dialog. If he can direct the theatergoer's imagination and 
interest, he has fulfilled his mission. The psychology of color is 
all-important in this respect, and we shall now show the manner in 
which certain colors upon the screen will give rise to certain emotions 
in the audience. 

We have found that by the understanding use of color we can 
subtly convey dramatic moods and impressions to the audience, 
making them more receptive to whatever emotional effect the scenes, 
action, and dialog may convey. Just as every scene has some definite 
dramatic mood some definite emotional response which it seeks to 
arouse within the minds of the audience so, too, has each scene, each 
type of action, its definitely indicated color which harmonizes with 
that emotion. 

The usual reaction of a color upon a normal person has been defi- 
nitely determined. Colors fall into two general groups. The 
first group is the "warm," and the second the "cool" colors. Red, 
orange, and yellow are called the warm or advancing colors. They 



Aug., 1935] COLOR CONSCIOUSNESS 143 

call forth sensations of excitement, activity, and heat. In contrast, 
green, blue, and violet are the cool or retiring colors. They suggest 
rest, ease, coolness. Grouping the colors in another manner we 
find that colors mixed with white indicate youth, gaiety, informality. 
Colors mixed with gray suggest subtlety, refinement, charm. When 
mixed with black, colors show strength, seriousness, dignity, but 
sometimes represent the baser emotions of life. 

As to the use of a single color alone, each hue has its particular 
associations. For example, red recalls to mind a feeling of danger, 
a warning. It also suggests blood, life, and love. It is materialistic, 
stimulating. It suffuses the face of anger, it led the Roman soldiers 
into battle. Different shades of red can suggest various phases of 
life, such as love, happiness, physical strength, wine, passion, power, 
excitement, anger, turmoil, tragedy, cruelty, revenge, war, sin, and 
shame. These are all different, yet in certain respects they are the 
same. Red may be the color of the revolutionist's flag, and streets 
may run red with the blood of rioters, yet red may be used in a 
church ritual for Pentecost as a symbol of sacrifice. Whether blood 
is spilled upon the battlefield in an approved cause or whether it 
drips from the assassin's dagger, blood still runs red. The intro- 
duction of another color with red can suggest the motive for a crime 
whether it be jealousy, fanaticism, revenge, patriotism, or religious 
sacrifice. Love gently warms the blood. The delicacy or strength 
of the shade of red will suggest the type of love. By introducing the 
colors of licentiousness, deceit, selfish ambition, or passion, it will 
be possible to classify the type of love portrayed with considerable 
accuracy. 

Proceeding to the other colors, orange is bright and enlivening; 
it suggests energy, action. 

Yellow and gold symbolize wisdom, light, fruition, harvest, reward, 
riches, gaiety; but yellow also symbolizes deceit, jealousy, in- 
constancy in its darker shades, and particularly when it is tinged 
with green. 

Green immediately recalls the garb of Nature, the outdoors, free- 
dom. It also suggests freshness, growth, vigor. 

Dark green, blue, violet, and indigo are cooling, quiet colors. 
They are tranquil and passive. They do not suggest activity, as do 
the reds and orange. Blue is suggestive of truth ("true blue"), 
calm, serenity, hope, science, also cold steel, melancholy (we have 
the expression "blue as indigo"). 



144 NATALIE M. KALMUS [J. S. M. P. E. 

Purple is a color which does not occur in the spectrum. It is a 
combination of warm red and cool blue. It will be aggressive and 
vital if the red predominates, or dignified and quiet if the blue over- 
balances the red. Purple denotes solemnity, royalty, also pomp 
and vanity. 

Magenta is the combination of purple and red. It is very dis- 
tinctly materialistic. It is showy, arrogant, and vain. 

The neutrals, white, gray, and black, while theoretically not in 
the category of colors, also stimulate very definite emotional re- 
sponses. Black is no color, but absorption of all color. It has a 
distinctly negative and destructive aspect. Black instinctively 
recalls night, fear, darkness, crime. It suggests funerals, mourning. 
It is impenetrable, comfortless, secretive. It flies at the masthead 
of the pirate's ship. Our language is replete with references to 
this frightful power of black black art, black despair, black-guard, 
blackmail, black hand, the black hole of Calcutta, black death (the 
devastating plague of medieval Europe), black list, black-hearted, 
etc. 

Even the poets recognized this symbolism. Shelly, in his dramatic 
Alaster tells how, 

"I have made my bed 

In charnels and on coffins, where black death 

Keeps record of the trophies won " 

The poet Keats, in The Prisoner of Chilian, says, 

"I, only, stirred in this black spot, 

I, only, drew the accursed breath of dungeon-dew." 

We are speaking a potent language to our audience when we make 
use of black. 

Gray suggests gray skies and rain. It is gloomy, dreary, and 
represents solemnity and maturity. From its complete neutrality 
and lack of any color or distinctiveness, it represents mediocrity, 
indecisiveness, inaction, vagueness. 

White reflects the greatest amount of light, it emanates a lumi- 
nosity which symbolizes spirit. White represents purity, cleanliness, 
peace, marriage. Its introduction into a color sublimates that color. 
For example, the red of love becomes more refined and idealistic as 
white transforms the red to pink. White uplifts and ennobles, 
while black lowers and renders more base and evil any color. To the 



Aug., 1935] COLOR CONSCIOUSNESS 145 

degree in which colors are lightened or darkened will the qualities 
that the color exemplifies be altered. 

Thus we see that all the colors in the spectrum speak their par- 
ticular language. The flush of anger, the vigor of a sun-tanned 
skin, the richness of gold velvet, the violet mystery of distant moun- 
tains, the serenity of blue sky these colors alone speak with more 
eloquence than could be described by words. 

The modification of a positive color by the introduction of another 
hue modifies the mental reaction to the degree of the intensity of 
that hue which is introduced. For example, a positive blue is a 
cool color, but to the extent in which a red hue is introduced, the 
coolness of the blue will be altered by the warmth of red. However, 
these complexities do not alter the basic principles of color or the 
general reactions which we have outlined. 

In the preparation of a picture we read the script and prepare a 
color chart for the entire production, each scene, sequence, set, and 
character being considered. This chart may be compared to a 
musical score, and amplifies the picture in a similar manner. The 
preparation of this chart calls for careful and judicious work. Subtle 
effects of beauty and feeling are not attained through haphazard 
methods, but through application of the rules of art and the physical 
laws of light and color in relation to literary laws and story values. 
In the first place, this chart must be in absolute accord with the story 
action. Again, it must consider the art, principles of unity, color 
harmony, and contrast. Again, it must consider the practical 
limitations of motion picture production and photography. The 
art director, however, in handling a color picture, must be forever 
mindful that the human eye is many times more sensitive than the 
photographic emulsion and many times greater in scope than any 
process of reproduction. Therefore, he must be able to translate 
his colors in terms of the process. 

When we receive the script for a new film, we carefully analyze 
each sequence and scene to ascertain what dominant mood or emotion 
is to be expressed. When this is decided, we plan to use the appro- 
priate color or set of colors which will suggest that mood, thus 
actually fitting the color to the scene and augmenting its dramatic 
value. 

We plan the colors of the actor's costumes with especial care. 
Whenever possible, we prefer to clothe the actor in colors that build 
up his or her screen personality. In a picture which we recently 



146 NATALIE M. KALMUS [J. S. M. P. E. 

completed, two young girls play the parts of sisters. One is viva- 
cious, affectionate, and gay. The other is studious, quiet, and re- 
served. For the first we planned costumes of pink, red, warm browns, 
tan, and orange; for the second, blue, green, black, and grey. In 
this way the colors were kept in unison with their film characters. 

One very important phase of making color pictures is the necessity 
of obtaining distinct color separation. The term "color separa- 
tion" means that when one color is placed in front of or beside 
another color, there must be enough difference in their hues to 
separate one from the other photographically. For example, there 
must be enough difference in the colors of an actor's face or costume 
and the walls of the set to make him stand out from the colors 
back of him; otherwise, he will blend into the background and 
become indistinguishable, as does a polar bear in the snow. If the 
colors are properly handled, it is possible to make it appear as though 
the actors were actually standing there in person, thus creating the 
illusion of the third dimension. Because of the general warm glow 
of flesh tints, we usually introduce the cooler tones into the back- 
grounds; but, if we find it advantageous to use warmer tones in the 
set, we handle the lighting so that the particular section in back of 
the actor is left in shadow. This gives a cool contrast to the faces, 
even though we have a general feeling of warmth in the room. When 
there are a number of players, all wearing differently colored cos- 
tumes, it is necessary to disregard those playing relatively unim- 
portant parts, and make the background in contrast to those whose 
action is most significant to this particular scene. 

It is important that the sets have interest and variety. They 
must not be flat. When the sets have depth it is much easier to 
introduce interesting shadows and colored lights for special effects. 
Unless the dramatic aspect dictates to the contrary, it is desirable 
to have all the colors in any one scene harmonious. Otherwise, we 
strike an unpleasant, discordant note. 

A point to be considered in set dressing depends upon one of the 
rules of composition in art. The law of emphasis states in part 
that nothing of relative unimportance in a picture shall be empha- 
sized. If, for example, a bright red ornament were shown behind 
an actor's head, the bright color would detract from the character 
and action. Errors of this nature must be carefully avoided. 

Color juxtaposition also plays a large part in the selection of 
colors for the screen. The effect of "color juxtaposition" is an 



Aug., 1935] COLOR CONSCIOUSNESS 147 

apparent change of hue when different colors are placed one over 
the other, or side by side. If two cards, one orange, the other blue- 
green, are placed side by side, the orange will appear more red than 
it really is, the blue-green more blue. Each color tends "to throw" 
the other toward its complement. In other words, the complement 
of orange is blue; therefore, the orange makes the blue-green appear 
bluer. When any two colors are placed together, the first emphasizes 
in the second the characteristics which are lacking in the first. 

It can readily be seen from this how exceedingly important it is 
to consider the movement in the scene in determining its color com- 
position because the juxtaposition of colors is constantly changing 
due to this movement. Quite a different problem from that of an 
artist, who paints a still scene where the characters remain in their 
set places, and whose color values, therefore, are not subject to 
frequently changing contrast. 

We must constantly practice color restraint. In the early two- 
color pictures, producers sometimes thought that because a process 
could reproduce color, they should flaunt vivid color continually 
before the eyes of the audience. This often led to unnatural and 
disastrous results, which experience is now largely eliminating. 

The synthesis of all these factors entails many conferences with 
directors, art directors, writers, cameramen, designers and others. 
Technicolor color directors, cameramen, and technicians act in a 
consulting and advisory capacity to the various studio departments 
during both the preparation and the shooting of the picture. 

Music, graphic art, and acting have now been united, and become 
one expression of more ultimate art. Now for the first time a perfect 
expression of the combined inspirations of producer, writer, artist, 
actor, and musician can be adequately presented to an audience. 
Color has touched the sound picture and it fairly lives. 



SOME PROBLEMS IN DIRECTING COLOR 
PICTURES* 



ROUBEN MAMOULIAN** 

Summary. A brief discussion of whether color in the motion picture is here to stay, 
pointing out that black-and-white was a convention that had to be accepted because of 
technical limitations at the beginning of the art. Had color -pictures been invented first, 
a black-and-white picture would now seem flat and inadequate although there are 
beauties in the unreal shadows that can not be denied nor be destroyed. The paper 
concludes with a few remarks upon the use of color for enhancing and emphasizing the 
emotional situations of a picture, and the effects to be achieved by carefully selecting 
the colors of the clothing or uniforms of the actors and of the backgrounds and lighting. 

No art has ever depended so much upon science as the art of 
motion pictures. In that sense it is truly the most modern of arts. 
It begins where science ends and it has a hard time, and not always 
a successful time, in artistically keeping up with the progress of the 
scientific and technical achievements that are taking place con- 
stantly in motion pictures. 

Seven years ago motion pictures were revolutionized by the advent 
of sound. Theretofore silent, the screen acquired the gift of speech. 
Today, as another result of scientific achievement, color comes to 
the screen, and to my mind, it is just as much a miracle as sound 
was. I should like to pay my most respectful tribute to those 
persons whose names one does not hear but who work in the silence 
and solitude of their laboratories. I refer to the scientists that 
compose the body of Technicolor, whose destinies are guided by 
Dr. Kalmus. 

The main question today is, "Will color last or will it not?" I 
have no doubt that color upon the screen is here to stay. I have 
also no doubt that there will be as much skepticism for the first 
few months in regard to color as there was in regard to sound. 

They say that what we do not have, we do not miss. No one 

* Presented at a meeting of the Technicians Branch of the Academy of Motion 
Picture Arts and Sciences, Hollywood, Calif., May 21, 1935. 

** Director of Becky Sharp, the first three-color feature motion picture produced 
in Technicolor. 
148 



DIRECTING OF COLOR PICTURES 149 

ever missed electricity until it came to replace oil and gas. No one 
missed dialog upon the screen while the screen was silent. However, 
let a dumb man, after thirty years of life, acquire the gift of speech; 
would he want to give it up and go back to his silence? Speech 
came to the screen and stayed victorious. Now, let a man with 
ailing eyes wearing black glasses through which the world looks 
gray, suddenly recover his sight, throw away his glasses, and see 
the luxury of color of the sky, the earth, and the flowers; would he 
ever want to go back to his black glasses? We never missed color 
upon the screen because the very art of the cinema was born black 
and white. It was a convention that had to be accepted. But 
once real color comes to the screen, we shall feel its absence as force- 
fully as we feel the absence of sound when looking at a silent film 
made some years ago. 

I do not mean to say that necessarily all the films will have to be 
in color, but certainly the great majority of them will be. As in the 
art of painting, while we admire and love black-and-white drawings 
and etchings, could we ever do without paintings? So far, the 
screen has been using a pencil ; now it is given a palette with paints. 

I do not want to be misunderstood. I do not want to imply that 
the black-and-white film is not beautiful, nor that the color-film 
completely displaces the black-and-white. As a matter of fact, 
the black-and-white has a beauty of its own that could never fade 
away. The very unreality of those pale shadows moving upon the 
screen, and that remote quality of a dream, constitute the attraction 
and the spell of the black-and-white film that could not be destroyed. 
There will always be room for certain subjects to be treated in terms 
of these fascinating gray shadows. But color comes to the screen 
now as a new spring to the earth. It comes as an inspiring and 
exciting gift, which opens new horizons of creation for the artist and 
enjoyment for the onlooker. 

I am stating this now not merely as a theoretical point, but as a 
result of an actual experience I went through recently. This ex- 
perience was directing Becky Sharp, the first full-length feature in 
color. That was a new and wondrous adventure. It had all the 
thrill and excitement of pioneering in a new field and discovering 
a theretofore unexplored fairyland. 

Color is one of the most powerful and fascinating attributes of 
nature. Imagine what the world would look like if you took color 
out of it. What would life be if we were forced to spend it among 



150 ROUBEN MAMOULIAN [J. S. M. P. E. 

sky, trees, flowers, and all things black, gray, and white? Having 
known the living joys of color, we should probably die of melancholia. 

Love of color and susceptibility to color are among the strongest 
instincts in human beings. If you want to discover the most organic, 
basic elements of the sophisticated human being of today, go to 
children and to savages. You will find that next to food, they love 
things of vivid color, which sparkle. That instinct is alive and strong 
in every one of us. 

In relation to motion pictures, our need for color has so far been 
ungratified. We accepted the situation just as we had accepted 
the fact of moving upon solid ground, until we learned to fly. But 
once color comes to the screen, we shall be unhappy without it. It 
brings a new terrific power to the screen. Our strongest impressions 
come through vision. So far, visually, we are dealing with light and 
shade and compositions upon the screen. Now we have an addi- 
tional element of color. This, not merely superficially to adorn 
the images in motion, but to increase the dramatic and emotional 
effectiveness of the story which is being unfolded to the spectator. 

Color, like all power, can be harmful and destructive when used 
badly; life-giving and creative when used well. Animals and 
human beings have always been and are unconsciously subject to 
the influence of color. How many times have you walked into a 
strange house and felt depressed because of the color of the wall- 
paper? How many times have you found consolation in the rich 
riot of shades of a gorgeous sunset? 

Apart from the pure pictorial beauty and the entertainment value 
of color, there is also a definite emotional content and meaning in 
most colors and shades. We have lost sight of that because, as with 
all important and inevitable phenomena, it has become subconscious 
with us. It is not an accident that the traffic lights of a city street 
today are green for safety and red for danger. Colors convey to us 
subtly different moods, feelings, and impulses. It is not an accident 
that we use the expressions "to see red," "to feel blue," "to be green 
with envy," and "to wear a black frown," Is it for nothing that 
we believe that white is expressive of purity, black of sorrow, red of 
passion, green of hope, yellow of madness, and so on? The artist 
should take advantage of the mental and emotional implications of 
color and use them upon the screen to increase the power and effective- 
ness of a scene, situation, or character. I have tried to do as much 
of that in Becky Sharp as the story allowed. To quote an example, 



Aug., 1935] DIRECTING OF COLOR PICTURES 151 

I would refer to the sequence of the panic that occurs at the Duchess 
of Richmond's ball when the shots of Napoleon's cannons are heard. 
You will see how inconspicuously, but with telling effect, the sequence 
builds to a climax through a series of intercut shots which progress 
from the coolness and sobriety of colors like gray, blue, green, and 
pale yellow, to the exciting danger and threat of deep orange and 
flaming red. The effect is achieved by the selection of dresses 
and uniforms worn by the characters and the color of backgrounds and 
lights. There is a little homecoming feeling in this for me, as the 
use of color and colored lights was one of my main joys and excite- 
ment in the theater. Surely, the effectiveness of productions like 
Porgy, Marco Millions, and Congai which I have done in the theater, 
would have been sadly decreased if I were forced not to use color in 
sets, costumes, and lights on the stage. 

Of course, in each art different subjects are expressed best through 
different forms. Undoubtedly, there are some stories that beg for 
color on the screen more than others do. Off-hand, a story of a 
historical period, when life and clothing were much more colorful, 
or stories with the backgrounds of countries like Spain or Italy, 
even of today, would ask for color more than some stories of our 
own modern age and civilization. The black-and-white films will 
still have their place upon the screen, but most assuredly as time 
goes by there will be less of them and more of the color pictures. 
For even though our life today is gray (and because of that), we have 
a great love and longing for color, is it not to be more attractive 
that women dress their bodies in beautifully shaded gowns and touch 
their faces with the subtle magic of a discriminating make-up? Is 
it not the same impulse that drives the gray and tired families of 
workingmen out to Sunday picnics, where there is a touch of blue 
sky, a green blade of grass, a tree, or a flower? 

Everything that is beautiful to the eye is a great gift to humanity. 
Color upon the screen is such a gift. The only danger of it that I 
can see during the first stages of the color picture, would be the 
danger of excess. Talking pictures did not avoid excess during the 
first months of their existence. There was too much talk and too 
much noise coming from the screen. The cinema must not fall 
into such another trap, and must not go about color as a newly-rich. 
Color should not mean gaudiness. Restraint and selectiveness are 
the essence of art. 



152 ROUBEN MAMOULIAN [J. S. M. P. E. 

DISCUSSION 

MR. K. MACGOWAN:* I'm going to start by repeating something I said at the 
Society of Motion Picture Engineers' luncheon yesterday. I'm not repeating 
it for my own benefit but because I want to say it to the members of the Tech- 
nicians Branch as well as of the Society. I feel a great deal of envy for you 
people who work on the technical end. As a producer I'm always running up 
against questions like these: Is it safe to do this story? Can I do this story as 
well and as uncompromisingly as it ought to be done? 

So often in the producing end we find that we are restricted by the public, or 
at least by what we think the distributors think the public thinks, and we end 
by doing a lot of things not nearly as well as they ought to be done. 

I noticed, however, in the first three months I was out here that the technicians 
always seem to do their job just as well as it could be done. They aren't up 
against the problem of what some picture owner thinks the public wants. No- 
body says that what Ray June is doing is over the public's head. Nobody says 
that the sound in One Night of Love is too good for the motion picture audience, 
nor does anyone say that Roberta is too well cut for the masses to understand. 

I envy the fact that you are always allowed to take as your motto, "perfection 
pays." 

Probably you are laughing up your sleeves because you find your equipment 
a little antiquated or you are up against financial difficulties, but you never have 
that horrible bug-bear "what the public wants," "the public won't stand for 
perfection." I know now that two major companies are debating whether or 
not they can make the greatest play in the English language, and that's a pretty 
disheartening idea. 

I want to tell you roughly the history of RKO's contact with Technicolor. 
About two years ago Merian Cooper persuaded Jock Whitney to make motion 
pictures in the new Technicolor process. At that time he was the head of the 
studio and I was lucky enough to be assigned the first picture in color. Then 
Mr. Cooper was taken ill. While he was away we determined not to make a 
long picture but a short. That was La Cucaracha. This fall when Mr. Cooper 
recovered and was ready to go to work again, he had to make two pictures in 
black and white, and again I had the good luck to be on the job and to do Becky 
Sharp. I'm boring you with this history only to give a bow to Merian Cooper 
as the father of three-color production on the screen. 

Another thing I want to say about technicians is that it seems to me they are 
wonderful people to deal with. I've found that true in the studio and particu- 
larly true at Technicolor. As soon as I went to work on La Cucaracha and 
Becky Sharp I came more and more into contact with these people and I found 
them quite as intelligent, quite as far-seeing as I had found them twenty years 
ago in Philadelphia when, as a motion picture editor, I first came in contact with 
Dr. Kalmus and his co-workers. I could name half a dozen men at Technicolor 
who have done wonderful work, not only in devising this new process but in 
cooperating with and understanding the rather screwy people connected with 
production. 

* Associate Producer, RKO-Radio Studios. Producer of Becky Sharp. 



Aug., 1935] DIRECTING OF COLOR PICTURES 153 

Before you see the reels of film that we have here to show you, I should like to 
point out one thing that seems quite significant to me. There is some resistance 
to color due to the fact that we discovered black-and-white photography first. 

Suppose there had never been black-and-white photography or black-and-white 
halftone reproduction. Suppose we had been used to color photographs and 
colored pictures for the past fifty or seventy-five years. Then if someone invented 
the black-and-white photography and black-and-white halftones, the result would, 
I am sure, be frightfully disappointing and definitely puzzling. We should have 
to translate all the tones almost as we translate a foreign language mentally 
when we hear someone speaking it. We should have to figure out mentally what 
actual color was represented by the gray of a face, the black of a tree, etc. I 
found somewhat the same effect after I saw a two-color picture, The Wax Museum. 
When a normal black-and-white picture came upon the screen it gave me a 
curious psychological shock and the thought, "What is this a painting in mud?" 
Experiences like this are going to beat down our instinctive resistance to color. 

One thing that is going to push color very far ahead is television. I was in 
the theater a good many years as a producer, and I saw the road destroyed by the 
movies. The silent screen was destroyed by the talking picture. Then 
the talking picture had to meet the competition of the radio. Now they tell us 
television is coming, and motion picture producers are beginning to worry about 
it. People can turn a little button and sit at home and be entertained, but they 
are going to get that entertainment in black-and-white for a good many years. 
Color television will come undoubtedly, but it will come late, and in the meantime 
the screen will be able to use color against the competition of television. There 
will be an added sense of vividness in the theater that will not be apparent upon 
the home screen. 



IMPROVEMENTS IN SOUND QUALITY OF NEWSREELS* 

J. A. BATTLE** 



Summary. The manner in which improvements in new sr eel sound quality were 
effected by coordinating the recording practices of the newsreel producers is described. 
A theater survey revealed that the projectionists reduced the fader setting for newsreels 
from one to five steps, or three to fifteen db., below the nominal fader settings for 
feature pictures to obtain the same volume level in the auditorium. By adherence 
to accepted recording practices in the studio newsreels are now being reproduced 
at the same fader settings as feature pictures, with a considerable improvement in 
sound quality. A brief outline is given of the preparation of the newsreel from news 
camera to theater. 



A "new deal" in newsreel sound quality was inaugurated on March 
4, 1935. Newsreel sound quality, it was generally recognized, was 
not keeping pace with the improvements in feature picture sound 
quality, and the increasingly greater contrast became so apparent 
that it was evident that some measures should be taken to eliminate 
this contrast. The recordists were cognizant of the reasons for the 
lack of a commensurate improvement, but as long as one newsreel 
had to be as loud as or louder than its competitor it was futile to ex- 
pect any marked improvement. The coordinated policy of record- 
ing at reduced levels adopted among the newsreel producers, which 
became effective with the release of March 4, resulted in a considerable 
improvement in sound quality; and with this as a nucleus further 
improvements are bound to follow. 

It was the original practice to reproduce the newsreels at slightly 
higher levels than the feature pictures, the theory being that the in- 
terest of the audience should be aroused and that the louder volume 
would enhance the value of the news. The reason for that seemed 
to be that the newsreel was apparently a lineal descendant of the 
town crier whose stentorian voice informed the vicinity at large of the 
news of the day. At first the louder newsreel volume was accom- 
plished by adjusting the fader settings in the theater in accordance 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Electrical Research Products, Inc., New York, N. Y. 
154 



IMPROVEMENTS IN NEWSREELS 155 

with a cue sheet, but later the increase in sound level was accom- 
plished upon the film itself. Gradually, this slight difference was 
increased to such an extent that the volume level of the newsreels in 
the auditorium sometimes became actually uncomfortable; so much 
so, that projectionists began to acquire the habit of automatically 
"pulling down" the fader from one to five steps, or three to fifteen db., 
when showing newsreels. 

In order to present to the newsreel producers convincing evidence 
that the method adopted to achieve loudness was defeating its own 
purpose, a survey of 150 theaters was made, unknown to the projec- 
tionists, by service engineers on their regular visits over a period of 
two or three weeks. Observations were made of the average fader 
settings of the feature picture and of the newsreel upon the same pro- 
gram, and the interesting figures shown in Table I were obtained. 

TABLE I 

Average of 
5 Newsreel 

Change from Average Feature Fader Companies 

Setting for Newsreel Projection (Per Cent) 

No change 6 

Down 1 step ( 3 db.) 21 

2(6") 25 

3(9") 30 

4 (12 " ) 14 

5 (15 " ) 4 

100 

It was evident that the higher recording level was being offset by 
the lower fader settings for the newsreels in the theaters. In order 
to obtain the increased volume level the recording equipment was 
operated beyond its optimal limits, contrary to the better judgment 
of the recordists, and light prints were released. The obvious solution 
to improve the quality was to adhere to accepted standards in re- 
cording and processing, or, in other words, to strive for feature pro- 
duction quality and return the control of the auditorium volume 
level to the proper place, i. e., to the projection room fader. 

The results of the survey were presented to the newsreel producers, 
and in subsequent discussions the advisability of correcting the con- 
dition was recognized. A conference was arranged at which the 
current release of each of the five newsreels was compared with a 
feature and a travelogue, and the comparative volume levels were 



156 J. A. BATTLE [J. S. M. p. E. 

measured with a volume indicator. The level of the newsreels ranged 
from +2 db. to +7 db., with reference to the level of the travelogue, 
which was arbitrarily selected as zero reference. The volume level 
of the feature measured 4 db. The zero reference level of the 
travelogue was considered suitable for newsreels, and it was agreed 
to adopt this level as a "yardstick." Prints of a section of the "yard- 
stick" were distributed among newsreel producers. With the re- 



PROJECTIONISTS 

BEGINNING WITH FIRST RUN RELEASE 

MARCH 4, 1935 

NEWSREELS OF ALL. COMPANIES WILL BE 
ISSUED AT LOWER VOLUME LEVEL 

MB* KM 

WARNING 

RUN THIS AND FUTURE NEWS- 
REELS ON NORMAL FADER 
STEP USED FOR FEATURE 
PICTURES 



FIG. 1. Notice distributed with initial release. 

duced level the projection room fader would remain unchanged for 
the news and the desired auditorium level would be obtained. The 
release date for the first news recorded according to the new stand- 
ards was selected as March 4. This date allowed sufficient time in 
which to transmit the news of the change to all projectionists by 
service engineers and by publicity in the trade papers. With this 
ground work laid, the March 4 release was distributed with the warn- 
ing notice shown in Fig. 1. 



Aug., 1935] IMPROVEMENTS IN NEWSREELS 157 

After a month's interval it was interesting to observe that the 
maximum variation in level of all the newsreels measured at a subse- 
quent conference was not more than ==2 db.; and, of greater impor- 
tance, there was a decided improvement in quality. Three of the 
reels were identical, one was 2 db. lower and another 2 db. higher. 
This close agreement has been verified by the records of the cue sheet 
of a New York newsreel theater in which all newsreels are shown in- 
tercut as a continuous program. The maximum volume adjustment 
between subjects was reported as 2 db., whereas formerly the varia- 
tion was as great as 8 db. Now that the benefits of improved re- 
cording have been realized there is not likely to be a recession to the 
old methods. There doubtless will be instances when the feature 
level is slightly lower than normal, which will necessitate a change 
in the fader setting for the newsreel, but the quality of reproduction 
will not be altered. The continued cooperation of the newsreel pro- 
ducers, combined with periodic conferences every six months or so, 
will tend to assure a continuance of the benefits attained. Further 
improvements in sound quality will most likely ensue in the footsteps 
of this first advance. 

The men in the field who gather the news are not affected by the 
change because, not having the same test and maintenance facilities 
available in the studio, nor the spare equipment, they must of neces- 
sity operate within conservative limits. They have to rely upon 
the home or district office to furnish their supplies, including film. 
A crew will use on an average of 50,000 feet of film a year, and some 
crews as much as 100,000 feet. They are constantly in touch with 
their local office, and receive practically all assignments from that 
source. In large centers such as New York and Chicago the crews 
are dispatched upon news assignments from the news desk, which is 
continually informed of the current events through the medium of 
the news teletype, telephone, etc. In some instances, a contact man 
accompanies the crew and interviews the "headline" celebrities. In 
more remote sections where the course of events is less hectic the 
crews develop their own stories, except in instances of happenings of 
national interest, such as disaster due to hurricanes, floods, or, for 
example, the recent dust storms. 

When completed, the stories are sent directly to the home office by 
air-mail; except when the story is not headline news: then the film 
is shipped by fast train. The cameraman reports the details of the 
story, including such items as camera angles and the number of feet 



158 J. A. BATTLE [J. S. M. P. E. 

shot. This report is attached to the film can and a duplicate is 
mailed at the same time. From it the newsreel editors can have a 
suitable script prepared for the commentator while the film is being 
developed. Advantage is taken of every means that will reduce the 
time interval between taking the story and reproducing it upon the 
screen in the theater. Each story is assigned an identifying number 
when reviewed in the negative by the news editors, who select and 
edit the material for release. A title and a short synopsis of the story 
are dictated by the editor, and a record is kept of all stories received. 
About 20 per cent of the stories sent in are used for release. All are 
catalogued and stored in a vault essentially as they are. The selected 
stories are edited and cut, and a dupe negative and a lavender print 
are made. The dupe negative is made as a matter of precaution and, 
on occasion, is used to double the printing capacity to expedite the 
release. The lavender "work" print is used for projection in scoring 
the comments, music, and sound effects if any are required. The 
commentator reads from a prepared script, and after a rehearsal or 
two the final sound negative is recorded. The original recorded 
sound may be used for a final release print or it may be re-recorded as 
an underlay when adding the comments. 

The preparation of the news begins one day prior to the release date, 
and an entire reel is ready to be printed within 18 or 20 hours. The 
news subjects are arranged in definite sequence prior to making the 
reel, and when the first half of the reel is scored prints are run off. 
The second half follows as quickly as possible, so that the release 
print is generally made up of two sections spliced together. For 
checking, a rush print is made from the sound negative and lavender 
print. The combined print has a negative of the picture, but this 
is satisfactory for the purpose. The entire process is so thoroughly 
checked during the assembly of the reel that it is not necessary to 
await the complete print for a final check. The completed reels are 
dispatched throughout the country according to a definite schedule, 
and if necessary delivery is accomplished by airplane to maintain the 
proper schedule in the event that there is some delay in making up, 
or in cases of specials or flashes. 

Special news stories of national importance are handled as 
"flashes" ; and in such cases, in order to eliminate the delay of await- 
ing "work" or "scoring" prints, the original negatives are used. In 
this way the stories are upon the screen within a few hours after the 
arrival of the negative. It is not at all uncommon in New York to 



Aug., 1935] IMPROVEMENTS IN NEWSREELS 159 

see a newsreel flash of the deciding touchdown of an important foot- 
ball game in the theater the evening of the same day. 

The news throughout the world is handled in about the same 
manner. In the key cities of the globe, London, Paris, Berlin, Rome, 
and Sydney, news events of world-wide importance are dispatched 
to the other five distribution centers by the fastest means of trans- 
portation. For Canadian and foreign distribution "dupe" nega- 
tives of the combined prints are employed. Speed is the essence of 
news, and the transmission of newsreels by television across seas and 
continents will probably be a reality in the future. 

The work of the sound department is continuous and not finished 
upon the release of the film. Reports that include the comments of 
the editor as well as the recording analyses are returned with every 
story, and by such means a continual check is available to the field 
crew upon the operation of their equipment. If, for instance, the 
density of the sound-track indicates a gradual change for two or 
three consecutive stories, the crew is informed to make the necessary 
adjustments. 

Prosaic facts such as these bear no intimation of the adventure 
and excitement of gathering news ; but the men who travel around the 
world to record the scenes and sounds know differently. Even in 
the more densely populated and civilized places they are likely to 
be on hand when the "fireworks" begin because they shoot anything 
and everything that will make news; and quite frequently sudden 
unforeseen happenings such as assassinations are recorded by the 
news cameras. The newsreel presents news as it actually happens, 
in a compact form unbiased by prejudices or opinions. 

DISCUSSION 

CHAIRMAN FRAYNE : How do you take care of the more or less inherent differ- 
ence of output between variable-density and variable-area films? 

MR. HUMPHREY: Each newsreel company was furnished with a portion of a 
print, so that their prints, run in their own systems, could be checked against the 
sample. 

CHAIRMAN FRAYNE: Was that done by changing the recording level or the 
transmission of the prints? 

MR. HUMPHREY: We left that to the newsreel studios. They do it in their 
own way, as they see fit. They make the check themselves and, as mentioned, 
do a pretty good job, because when the prints get to the theaters they seem to be 
quite uniform. 

CHAIRMAN FRAYNE : It is certainly encouraging to learn that the various news- 
reel companies can get together and bring about such an improvement. I could 



160 J. A. BATTLE 

not help but think, as I listened to Mr. Humphrey, that if the studios of Holly- 
wood could do likewise, the public would be given a lot of relief. 

At the present time we go into a theater and hear a feature, or a reel of one, the 
sound volume of which is very comfortable. We hear and understand perfectly. 
Then all at once the picture changes to a Mickey Mouse, or a newsreel, or to some- 
thing else. Suddenly there is a tremendous change in the output of the horns, 
and a great deal of discomfort is caused among the patrons, which must certainly 
be reflected eventually at the box-office. 

If the labors of the Sound Committee can accomplish half as much good as 
has been done with the newsreels, a great job will have been done. 



A DYNAMIC CHECK ON THE PROCESSING OF FILM FOR 
SOUND RECORDS* 



F. G. ALBIN** 

Summary. The limitations of the sensitometric method of controlling film de- 
velopment as actually affecting the sound recording are reviewed. An auxiliary 
test designed to check the quality of the development technic is outlined. In this test, 
the over-all photographic process, including the light-valve circuit which controls the 
exposure of the negative, and the photoelectric cell which receives the current impulses 
in accordance with the transmission of the positive film, are considered as an electrical 
unit. This unit is analogized to a transducer in an electrical transmission circuit 
and, as such, its fidelity of transmission can be determined by comparing the electrical 
output with the electrical input. In practice this is accomplished by applying a 
constant amplitude of electrical input to the light-valve circuit at values of valve 
spacing selected to allow the peak exposures to extend to the limits of the exposure 
regions that are to be used in the recording. The resultant projected levels for each 
value of valve spacing are measured with a volume indicator on the projection amplifier. 
If the readings obtained are all the same, true reproduction over the exposure range 
used is indicated. 

In the majority of sound recording and reproducing systems as 
used today, photography on film is used to attain the required time- 
delay feature. The technic of the photographic process greatly 
affects the quality of reproduction and continuous effort is expended 
to improve the conditions. The control of the photographic process 
has developed into a science, but the popular control methods when 
reduced to their simplified practical form are incompetent to provide 
fundamental data. 

Considering the recording and reproducing system as a whole, the 
photographic portion, including the light- valve circuit which controls 
the exposures, and the photoelectric-cell circuit which receives the 
corresponding current impulses as a result of the exposures, may be 
analogized to an electrical transducer in an electrical circuit. This 
"transducer" (Fig. 1), as an electrical network, has as its terminations 
the circuits of the light-valve and the photoelectric cell. It may be 
characterized by the following features: 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** United Artists Studios, Hollywood, Calif. 

161 



162 



F. G. ALBIN 



[J. S. M. P. E. 



CO Time delay (function of system). 

(2) Ground-noise (inherent noise generated within the "transducer"). 

(3) Limit to input level (usually measured with reference to the ground-noise 

level). 

(4) Distortion (not desirable). 

(a) Wave-form. 

(6) Amplitude distortion as a function of time. 

(c) Frequency distortion (principally introduced by the mechanics of 
the system). 

(5) Transmission efficiency. 

(a) Absolute efficiency, i. e., ratio of output to input power, is relatively 

unimportant. 
(6) Discrimination as to frequency, relatively important. 

(6) Limit to output level (usually measured with reference to ground-noise 

level). 



c 




( 


,- > 








r ~ " 












"1 


i 












i 


HICKOPHONZ 












i 


i r-| 1 

KCCOXDIHG \ L y 

fl 




\ 


PR 

\ 


VTIN6 


fl" 





NfG. Xf*>$(4? 








^ \ 


posm 


* 










| 




iIoH 


H EG XT IV Z 


_ 






]U 


\z 


. 


PfVfi-OPMeN 


r~ 








EyELOFMC. 


V7 j 






( 


^ . 




G 





* TRANSDUCER 

FIG. 1. Analogy of the photographic portion of the recording and re- 
producing system to an electrical transducer. 

These characteristics can be measured electrically at the termina- 
tions more easily than by an accumulation of separate tests covering 
individual steps within the "transducer." 

The volume latitude available for reproducing any one frequency 
is limited by the fidelity of transmission at high output levels and by 
the ground-noise at low levels. An indefinite number of factors 
affects the fidelity of transmission through the "transducer," and some 
of the more important ones are listed in their conventional 
classification : 



Light-valve coefficient. 

(a) Huyghen's effect upon light passing through a small aperture. 



Aug., 1935J 



CHECK ON PROCESSING OF FILM 



163 



(&) Reciprocity law failure due to high intensity and short exposure 
time. 

(c) Resolving power variation with density, development time, etc. 

(d) Spectral composition of exposing illumination. 
(2) Negative gamma. 

(a) Choice of film emulsion. 

(&) Negative exposure. 

(c) Negative development. 

(1) Type and age of developer. 

(2) Speed of machine. 

(3) Etc. 




FIG. 2. Over-all graphical construction with H&D curves. 

(5) Printer coefficient. 

(a) Spectral composition of exposing illumination. 
(6) Etc. 

(4) Positive gamma. 

(a) Choice of film emulsion. 

(6) Positive exposure (printer point). 

(c) Positive development. 

(1) Type and age of developer. 

(2} Speed of machine. 

(5) Etc. 

(5) Projection factor. 

(a) Photoelectric cell type. 



164 F. G. ALBIN [J. S. M. p. E. 

(1) Gas. 

(2) Window size. 

(b) Optical system proportions. 

Of all the factors listed, negative and positive development are 
larger and more important. In practice they are subject to varia- 
tions which, in turn, greatly affect the others. To control this de- 
velopment the sensitometric method was adopted. The Hurter and 
Driffield gamma used was defined as the slope of the straight-line 
portion of the plot of density against the logarithm of relative ex- 
posure. The common methods of applying the exposures and mea- 




FIG. 3 (a). Over-all recording characteristics (theoretical). 

suring the densities produce conditions that differ greatly from actual 
operating conditions. Gamma, to be determined in this manner, 
must be augmented by the factors and coefficients given above. 

If gamma were used in a broader sense to respect exposures as 
applied by the light-valve and densities as recognized by the photo- 
electric cell, the various coefficients would be absorbed in the terms 
negative and positive gamma. With gamma used in this broader 
sense it holds true that the photographic requirement for true repro- 
duction, where the product of negative and positive gamma is unity, 
must be met. The requirement for recording is usually expressed as 



Aug., 1935] 



CHECK ON PROCESSING OF FILM 



165 



for 

in which 



T p = 



7 P = positive gamma. 

7n = negative gamma. 

T p = positive transmission. 

E a = negative exposure. 




FIG. 3(6). Over-all recording characteristics (practical). 

Several values of over-all gamma products are shown in Fig. 3 (a), and 
it can be seen that for linear reproduction the product must be unity. 
In practice, it has been found necessary to allow the exposures of 
the negative and the positive to extend beyond the straight-line por- 
tions of the negative and positive curves. This is shown in the graphi- 
cal construction using H&D curves (Fig. 2.). In the theoretical 
plot the logarithmic mean exposures on the positive and negative 
curves are used as mean exposures. In the practical illustrations, 
positive mean exposures are determined by the choice of over-all 



\ 

166 F. G. ALBIN [J. S. M. P. E. 

gamma products. The negative exposure is governed by the avail- 
able illumination, and where possible is usually the logarithmic mean 
of the straight-line portion. 1 Whenever the exposures are not con- 
fined to the straight-line portions, the H&D gamma product loses its 
significance. It is necessary to consider the term of density gradient 
instead of H&D gamma. This term, density gradient, as has been 
defined, 2 is the slope at any point along the curve, and is an instan- 
taneous value. The use of the negative and positive density gra- 
dients instead of H&D gamma values in the above equation makes it 
applicable to types of recording wherein the exposures are allowed to 
extend beyond the straight-line portions and on to the toes or 
shoulders of the H&D curves. The equation then becomes 



where G n and G p are the negative and positive gradients, respec- 
tively. 

The curves for various over-all gamma products which were shown 
involve only the straight-line portions of the H&D curves, and as 
such are theoretical. When replotted (Fig. 3(&)) using gradients, 
these give actual conditions as obtained in practice, and include all 
the coefficients involved. The several curves indicate that for any 
degree of development of negative and positive there is a resulting 
over-all curve which has a point of inflection about which, for a 
limited amplitude, the curve is fairly straight. Thus, linear repro- 
duction may result with the development varied over a wide range, 
provided that for each over-all condition the mean projected positive 
transmission is adjusted to this point of inflection and the amplitude 
is restricted to the portion of the curve that is nearly straight. The 
gradient product over this region is approximately unity. In these 
curves, however, negative exposure has not been adjusted by the 
method previously mentioned to permit maximum valve opening. 
Fig. 4 shows one of the curves replotted to allow the mean valve 
spacing to produce a projected transmission equal to the optimal 
point P, and permits maximum valve opening with full modulation. 
Each curve could be similarly treated. The resulting plots would 
still show that as the over-all gamma product departs from unity the 
limits of the unity gradient region in terms of T 2 and T\ would be 
reduced, and that P would be greater with larger values of over-all 
gamma products. Correspondingly, with smaller values of over-all 
gamma products T 2 and TI would be reduced and P would be smaller. 



Aug., 1935] 



CHECK ON PROCESSING OF FILM 



167 



As applied to this over-all curve, the maximum T 2 and minimum 
TI values of projected transmissions that can faithfully be used in a 
cycle of modulation are determined by the extremes of this unity 
gradient product region. With maximum allowable amplitude, the 
peak values of the cycle would correspond to T 2 and T\ t but for lesser 
values of modulation the peak transmissions of the cycle would not 
cover the entire range. This allows the mean transmission of the 
cycle of modulation to be adjusted as desired to increase or decrease 
the output volume, without introducing distortion, provided the peak 




VALVE SPACING. 

FIG. 4. Over-all recording characteristics (dynamic). 

values do not extend beyond T 2 or 7\. This practice is usually re- 
ferred to as cueing. 

The projected level of the positive as measured with the volume 
indicator is the difference between the maximum and minimum peak 
values of the cycle being reproduced. The ground-noise for this 
cycle is determined by its mean transmission. For sine- wave forms 
the volume latitude, or the modulation of the positive, becomes 



168 



F. G. ALBIN 



[J. S. M. P. E. 



Using the values of volume latitude as obtained from the actual 
curves (Fig. 4) of transmission vs. exposure and plotting them against 
their corresponding over-all gamma values, an optimal development 
can be found (Fig. 5) . The over-all gamma product is used to iden- 
tify each development instead of the over-all gradient product, be- 
cause the latter, as has been shown, is approximately unity for all 
development conditions throughout the true reproduction range. 

In determining these curves so far the shape has been the result 
of selecting an over-all gamma. Now, if we can determine the shape 
by some other means, it follows that the over-all gradient can be ob- 
tained. The dynamic test is designed to accomplish this determina- 
tion and provide a simple and practicable means of recognizing a unity 
gradient product. 




0M0MC2 6 AM MA 

(COR freer r^AMSf 
FIG. 5. Curve of optimal development. 

THE DYNAMIC TEST 

If a very small amplitude of sinusoidal modulation is applied to 
the light-valve and a predetermined mean spacing is used, a certain 
projected level with its mean transmission results. By maintaining 
a constant modulation amplitude and setting the mean spacing at 
several points, a corresponding projection level for each point can be 
measured with a volume indicator. The volume indicator readings 
will be proportional to the slope of the over-all curve at the point of 
mean valve spacing. By exploring the available exposure range, 
the complete over-all curve of projection transmission vs. light- 
valve exposure can be constructed. The value of over-all gradient 
for the portion of the curve between any two mean exposure values, 
E n i and E n2 , is determined by the corresponding volume indicator read- 
ings as shown in the following equations: 



Aug., 1935] CHECK ON PROCESSING OF FILM 169 

EQUATION FOR DETERMINING OVER-ALL GRADIENT FROM VOLUME INDICATOR 

READINGS 

Let 

7n = negative gamma. 

TP positive gamma. 

G n = negative gradient. 

G p = positive gradient. 

A = projected amplitude. 

R = volume indicator readings (db). 

For negative and positive exposures on the straight-line portion of 
the H&D curves 

For any negative and positive exposure on the H&D curves 

1 p = .A. .tin n P 

Let 

C/nGp = G 

Taking the first derivative 

- = KGEn G ~ l 

Let the ratio of the projected amplitudes for mean exposures E ni 
and E n2 be 

A* 

A, 

Then with constant amplitude of negative modulation 



A 1 dT pl \E n 



But volume indicator readings are in decibels, whence 
Substituting 



- R! = 20 log 2 



R t - Ri = 20(G - l)log - 

Enl 

In the practical daily application of this test method (Fig. 5). 



170 F. G. ALBIN 

where optimal conditions have been determined from a more com- 
plete investigation, only three values of mean valve spacing are 
chosen, as it is evident that an over-all curve with three points of the 
same slope must be a straight line. The amplitude is fairly high to 
give a projected level considerably higher than the ground-noise 
level. The peak values of these spacings just cover the maximum 
exposure range used in recording during average conditions. The 
frequencies usually employed are 1000 and 5000 cps. The projected 
levels of the three exposures for each frequency when measured with 
the volume indicator, if flat or equal, show that the recording is not 
exceeding the straight-line portion of the over-all curve. A com- 
parison of the projected levels of the two frequencies is a measure 
of the resolving power of the photographic portion. If not flat, the 
test indicates that adjustments must be made. The nature of the 
adjustments is determined by the individual conditions. In general, 
the attenuation of projected levels at low transmissions is caused by 
(1) low negative density, (2) low positive transmission, or (3) high 
over-all gamma. Also, the attenuation of the projected levels at 
high transmissions is caused by (1) high negative density, (2) high 
positive transmission, or (3) low over-all gamma. In more than 1000 
tests, this method of testing has apparently faithfully indicated the 
film development condition where audible tests verify the indications 
given by the dynamic test. 

REFERENCES 

1 MACKENZIE, D. : "Straight-Line and Toe Records with the Light- Valve," 
/. Soc. Mot. Pict. Eng., XVII (Aug., 1931), No. 2, p. 172. 

2 JONES, L. A. : "On the Theory of Tone Reproduction with a Graphic Method 
for the Solution of Problems," /. Opt. Soc. of Amer., 4 (1920), p. 420. 



ENGINEERING TECHNIC IN PRE-EDITING MOTION 

PICTURES* 

M.J.ABBOTT** 

Summary. The use of engineering technic in determining the screen value of 
each scene in a motion picture before filming is described, and the methods used for 
control during the process of production are explained. 

Pre-editing motion pictures is nothing more than the application 
to motion pictures of engineering principles used in other lines of 
endeavor. In setting out to paint a picture, an artist first deter- 
mines its size. He then outlines his subject. He does not begin to 
paint his picture without first knowing the relationship each detail 
bears to the others. The same is true in constructing a building. 
The architect first determines the size and type of building desired 
by his client, and then applies his artistry in designing. After these 
principal facts have been determined, he applies his technical methods 
of construction. He uses principles which are based upon past ex- 
perience to determine the foundation necessary and other technical 
details that will make the completed building both artistic and prac- 
ticable. 

Motion pictures, being made exclusively for their commercial value, 
have their limitations as to size and the amount of money that can be 
derived from renting the prints. As the possible return upon the in- 
vestment is limited, the cost of the motion picture must of necessity 
be limited. Although the cost of an article usually has some bearing 
upon the possible return upon the investment, excessive costs will not 
lead to excessive returns unless the quality of the article has been en- 
hanced commensurately by the expenditure. 

The financial structure of the motion picture industry, established 
by its competitive industrial standing, is such that it not only deter- 
mines the possible return upon the investment, but also limits the 
length of the pictures. To fulfill conditions existing within the indus- 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RKO Studios, Hollywood, Calif. 

171 



172 M. J. ABBOTT [J. S. M. p. E. 

try, exhibitors are compelled to have at least two performances a 
night in order to operate their theaters profitably. Thus, the play- 
ing time for each show is naturally limited. As the exhibitor is com- 
pelled to present a balanced program to his patrons, experience hav- 
ing established the newsreels, comedies, educational features, etc., 
the playing time of a feature picture is limited to approximately one 
and one-half hours. The studios must construct their pictures to 
fulfill these requirements. 

A general impression exists that making a motion picture is 
entirely a matter of creative artistry and, as such, must not be hamp- 
ered by anything of a technical nature. It is not the purpose of this 
paper to attempt to prove or disprove this theory. We shall grant 
that it is true in part and that the creator's mind should not be an- 
noyed with technicalities while creating. However, the results of 
the creator's efforts become purely technical as soon as completed, 
and their value, technically or commercially, is determinable by com- 
parison with past experience. 

The principle of pre-editing is not to curb the creative mind by 
compelling it to consider the technical points of a picture while work- 
ing upon a story. It is a means of determining the motion picture 
value of the creator's efforts. Many question the possibility of com- 
puting the value of a story. This impression is not based upon facts, 
as story values have been measured since the inception of motion 
pictures. The only difference between pre-editing and the method 
that has been used in the past is that we do not wait until the story is 
upon the film and reaches the cutting room to find the errors, and then 
hope to edit the story by having the cutter remove the surplus or poor 
scenes and reduce the picture to commercial length. It can, there- 
fore, be seen that the principles used in pre-editing are not new, but 
are merely the application of these known principles in determining 
the value of the story before rather than after the picture has been 
made. 

The first principle to be considered in pre-editing is the type of 
picture to be made. These types are as follows : 

(1) Drama (5) Farce 

(2) Melodrama (6) Musical Revues 

(3) Comedy Drama (7) Musicals 

(4) Comedy (8) Westerns 

The determination of the type sets the tempo in which the picture 
will be made. The story must be timed accordingly. 



Aug., 1935] PRE-EDITING MOTION PICTURES 173 

The first step in determining picture value is to read and study the 
entire script; after doing which each scene of the script is read and 
studied individually, as though it were a complete picture. The 
action as outlined by the writer is studied and allowance made for the 
footage necessary to place each of the individual scenes upon the 
screen. The basis used in determining the necessary footage is that 
of the presently used cutting principles. The dialog of the scene is 
timed by reading the lines in the tempo of the class of picture to be 
made, allowance being made for the simultaneous occurrence of dia- 
log and action. 

The next step is to determine the value of each scene as a motion 
picture, namely, what portion of the scene, as outlined, will be given 
to the audience through the eye, and what portion through the ear. 
This is arrived at by timing the dialog in the scene which is not cov- 
ered by action. The scenes are then computed by episode and se- 
quence, and summarized for the entire story. The results are fur- 
nished to the writer, producer, and director for their guidance in re- 
writing the story, to eliminate or correct weak spots due to excessive 
dialog, and also to visualize the relative value each scene bears to 
the entire picture, and to regulate the length of the picture to meet 
the commercial requirements of the exhibitor. 

During the shooting of a picture, a production control record is 
kept, and the actual time of shooting each scene is compared with 
the estimated value of the scene as conceived by the writer. This 
control is arrived at by comparing the script notes made by the 
company script clerk, who times the footage of each scene during 
the process of shooting. As the various episodes and sequences of the 
story are completed and placed in ' 'rough cut" by the cutters, they 
report to the production control the amount of footage of each take 
made by the director that has been used and placed in "rough cut." 
Upon receipt of this information, and upon comparing it with the pre- 
estimate, it can be determined whether or not the picture is being shot 
in the tempo of the class of picture desired. 

Daily Production Reports are furnished to the producers and direc- 
tors showing the status of their picture, as to the shooting schedule, 
the quantity of film used, and the tempo in which the director is 
actually shooting the picture. As the story value of the picture in 
the pre-estimate is based upon the correct tempo of the type of pic- 
ture to be made, by comparing the actual shooting with the estimated 
time the producer is informed as to the tempo of each scene and se- 



174 M. J. ABBOTT 

quence, and as to whether or not it is shot too slow or too fast, so 
that when the picture is completed it will not be slow in spots and 
fast in others. 

When the picture is finally completed and ready for release, the 
actual takes used in the picture are compared with those shot by the 
director and the cost of those not used (the out-takes) is determined 
by the time spent in making them. Some of the benefits to be de- 
rived from pre-editing are as follows : 

(1) Eliminates waste due to over-shooting. 

(2) Shortens shooting schedules. 

(3) Saves time of company and executives in projection room checking film 
which never reaches the finished picture. 

(4) Prevents distorting the story by endeavors to cure defects after shooting. 

(5) Allows the story as written to reach the finished picture without mutila- 
tion. 

(6) Allows the judicious and profitable spending of money. 

(7) Hence the improvement in quality of product. 

(8) Finally, cleans up the cutting room floor. 



CHARACTERISTICS OF THE PHOTOPHONE LIGHT- 
MODULATING SYSTEM* 



L. T. SACHTLEBEN** 

Summary. Some of the requirements of the ideal light-modulating system for 
recording sound upon film are discussed, together with features of design and certain 
operating characteristics which should be incorporated into and exhibited by such a 
system. The Photophone light-modulating system is outlined, its general adapta- 
bility to the production of a wide variety of sound-track types reviewed, and its prac- 
tical conformity to the ideal requirements pointed out. 

Any practicable light-modulating system used in recording sound 
upon film, by either the variable- width or variable-density method, 
should be designed with certain ideal requirements in mind, and 
should be made to fulfill those ideal requirements as closely as pos- 
sible. Although the variable-width and variable-density sound- 
tracks are quite different in character, the apparatus used in making 
them is essentially the same, in that both employ long, narrow, il- 
luminated slits focused upon the moving film by optical means. 

In variable-width recording this slit image is modulated by varying 
the length of its illuminated portion so as to produce upon the 
finished film a strip of constant density and varying width, constitut- 
ing a wave geometrically perceptible to the eye, whose contour is a 
graphic picture of the pressure or velocity component of the sound- 
wave at the microphone diaphragm. The variable-width sound- 
track is essentially an oscillogram (Fig. la). 

Modulation of the slit image in variable-density recording is such 
as to vary the exposure from point to point along the length of the 
sound-track, keeping the slit image at all times uniformly illuminated 
along its length. The resulting sound-track varies in density from 
point to point along its length, its transmission varying, ideally, in 
linear relation to the pressure or velocity component of the sound- 
wave at the microphone diaphragm. Although the variations of 
density are apparent to the eye, the transmission of the track, or 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RCA Manufacturing Co., Camden, N. J. 

175 



176 L. T. SACHTLEBEN [J. S. M. P. E. 

its power to transmit light, is not geometrically apparent to the eye 
as in the variable-width sound-track (Fig. Ib). Since exposure is 
proportional to the product of image illumination and the time during 
which the emulsion is submitted to the action of the light, variable- 
density records may be made by varying the exposure time and keep- 
ing the illumination constant, or by varying the illumination and 
keeping the exposure time constant. 

In so far as the character of the slit and optical system are con- 
cerned, departure of the sound record from ideal perfection, through 
attenuation of the high frequencies, introduction of harmonic dis- 




FIG. 1. (a) Double-edged or bilateral variable-width sound- track; (b) 
variable-density sound-track. 

tortion, and so forth, is largely due to departure of the system from 
the following ideal conditions: 

(7) The slit image should be of infinitesimal width. 

(2) The slit image should be fixed in width. 

(3) The illumination should be perfectly uniform along the length of the slit 
image. 

(4) The exposure should be adequate to produce the densities required by the 
type of sound-track used. 

(5) The stray light ratio should be zero. 

In practice, recorder slit images are of finite width, and give rise 
to the slit effect, the magnitude of which can be kept small only by 
keeping the width of the slit image very small. The slit effect in 
variable- width recording has been analyzed by Cook 1 and Foster, 2 
the former having determined the ideal superior limit of the effect, 
and the latter having analyzed its extent under the conditions of 



Aug., 1935] THE PHOTOPHONE SYSTEM 177 

practice. Briefly, it is the introduction of harmonic distortion into 
a recording and the attenuation of the fundamental, due to the fact 
that the width of the recording slit image is an appreciable fraction 
of the wavelength (upon the film) of the recorded frequencies. In 
variable-density, the slit effect introduces only attentuation, and 
does not give rise to harmonic distortion. 

The recorder slit image is fixed in width in variable- width work, 
and in variable-density work where the image illumination is varied, 
as in glow-lamp recording. The image varies in width in variable- 
density work where the illumination is constant and the exposure 
time is varied. 

Any very great departure from uniform illumination along the 
length of the slit image will cause distortion in either variable- 
width or variable-density work. In variable-width, departure from 
uniform density across the sound-track will cause the transmission 
to be no longer linearly related to the amplitude of the impressed 
signal ; and in variable-density, non-uniformity in this respect may 
cause longitudinal portions of the sound-track to depart seriously 
from linearity. These distortions will arise in variable- width work 
if the density of one edge of the negative drops sufficiently to allow 
appreciable printing through, with attendant change of wave-shape ; 
and in variable-density work if non-uniformity of slit image illumi- 
nation is such as to cause a portion of the track to be recorded off 
the linear portion of the over-all curve. 

The fourth point, calling for adequate exposure, is more or less 
obvious. In variable-density recording by means of a glow-lamp, 
the classical straight-line recording scheme is abandoned in favor of 
toe recording, on account of limited illumination, and the light-valve 
has been adopted for straight-line recording where large exposures 
and large variations of exposure are required. 

A low ratio of stray light to useful light is important in variable- 
width work, in which the exposed portion of the track varies in 
width and the unexposed portion must receive no exposure, in order 
to assure good output at high densities and to keep down attenuation 
at high frequencies due to irradiation. 

Aside from approaching these ideal requirements, a practical 
light-modulating system should fulfill additional requirements of 
great practical importance. A practicable system should be simple 
to adjust and operate, and should be stable in adjustment even when 
subjected to serious overloads. It should be rugged in construction, 



178 L. T. SACHTLEBEN [j. s. M. P. E. 

protected against dust, and readily cleaned and kept in condition. 
The system should be capable of being checked in a simple manner 
as to condition and performance, or monitored, continuously during 
recording. Optical efficiency and the quality of optical correction 
should be high, and the system should incorporate a ground-noise 
reduction feature, to keep the ground-noise to signal ratio low at low 
recorded amplitudes. Power requirements should be reasonable, 
and the frequency response sufficiently wide to allow music and speech 
to be recorded with natural fidelity or life-like quality. 

THE PHOTOPHONE LIGHT-MODULATING SYSTEM 

Before discussing the features of the Photophone light-modulating 
system in detail, the more important ones will be presented in outline : 

(1) High illumination of the slit image makes it possible to use slow, fine- 
grained emulsions of high resolving power. 

(2) The slit image is fixed in width. 

(5) The slit image is 0.00025 inch wide, practically eliminating the slit effect. 

(4) The movements of the recording beam are sufficiently apparent to the 
eye to constitute an effective monitoring system. 

(5) All parts are rugged, thoroughly protected, capable of recording ampli- 
tudes well beyond the limit set by the width of the sound-track before over- 
loading, and will withstand large overloads without injury. 

(6) A ground-noise reduction circuit is built into the Photophone recording 
galvanometer. 

(7) The frequency response is essentially flat from to 9000 cycles. 

(8) The system is capable of recording three distinct types of sound-tracks, 
including the variable-density. 

(9) The images of the recording aperture and of the slit are both formed by 
special achromatic lenses. 

Construction and Operation (General}. In its essential form the 
new Photophone light-modulating system does not differ from 
the earliest variable- width recording systems, closely adhering to the 
arrangement described by Hardy in 1928. 3 Fig. 2 is a schematic 
diagram, in the horizontal plane, of the Photophone system, the 
operation of which may be briefly described as follows : the condenser 
lens b and the achromatic lens d image the lamp filament a upon the 
vibrating mirror e. Aperture c lies close to the condenser b, and is 
imaged sharply upon the slit g by the achromatic lens d. The 
condenser lens / images the mirror e upon the special corrected 
achromatic objective h, which images the slit g sharply upon the 
film. In operation, 4 the mirror e vibrates about an axis in the plane 



Aug., 1935] THE PHOTOPHONE SYSTEM 179 

of the figure and intersecting both optical axes in their common inter- 
section point. This moves the image of the aperture c up and down 
in the plane of the slit g. 

By making the aperture c in the form of an isosceles triangle with 
its base horizontal (parallel to the slit), its image will appear upon 
the slit g as shown in Fig. 3. As the aperture image moves up and 
down, the length of the illuminated portion of the slit will be in linear 
relation to the angular displacement of the mirror e. The variable- 
width sound-track thus produced will, in the print, have the general 
appearance indicated. It will readily be appreciated that in the 




Screen 

FIG. 2. Diagram of the Photophone light- 
modulating system. 

variable-width track the recorded wave is geometrically perceptible 
to the eye, its contour being a graphic picture of the pressure or 
velocity component of the sound-wave at the diaphragm or ribbon 
of the microphone. In fact, with the variable-width type of record, 
"faithfulness of reproduction is primarily a matter of reproducing a 
geometric form." 5 

If recordings are made in the manner just illustrated, the average 
transmission of the resulting sound-track will be constant and equal 
to 50 per cent, whether the modulation be high or low. When 
modulation is low it is desirable that the average print transmission 
be low also, in the interest of a low ground-noise to signal ratio. 
This end is attained 6 in the Photophone optical system by elec- 
trically biasing the position of the mirror e (Fig. 2) until only the 






180 L. T. SACHTLEBEN [J. S. M. P. E. 

tip of the triangular image falls upon the center of the slit (Fig. 4). 
When no signal is impressed upon the amplifier an extremely narrow 
transparent line occurs down the center of the printed sound-track. 
When a signal is impressed upon the amplifier, a special ground-noise 
reduction circuit causes the image of the triangle to rise propor- 
tionately to the envelope of the impressed signal until the average 
transmission of the track is 50 per cent, thereby increasing the 
average transmission of the print to just the extent necessary to 



IMAGE OF RECORDING 
APERTURE ON SLIT 




NATURE OF 
DOUBLE EDGED 
OR BILATERAL 

SOUND TRACK 



FIG. 3. The variable-width sound-track with- 
out ground-noise reduction. 

accommodate the wave without overshooting. In this manner, the 
ratio of ground-noise to signal is kept low at all times. The operation 
of the ground-noise reduction system will be described in more 
detail later. 

The adjustments of the high-fidelity recording optical system are 
relatively simple and uncritical (Fig. 5) . The mechanical slit, which 
is 0.0015 inch wide by 0.420 inch long, is fixed in permanent adjust- 
ment at the factory and presents no adjustment problems whatever 
in the field. It is so located in the optical system as to be entirely 
dust-proof, and with ordinary use and care will never be liable to 
injury or require cleaning. 



Aug., 1935] 



THE PHOTOPHONE SYSTEM 



181 



The exposure lamp is set in a universal mounting, and can be 
simply and quickly adjusted so that the image of the filament falls 
centrally upon the galvanometer mirror, which, incidentally, is 
covered with a plane glass window, tilted to introduce no stray light 
and designed to protect the galvanometer from dust and mechanical 
injury. 

Like the slit, the triangular recording aperture, which is imaged 
upon the slit, is set in a dust-proof mounting. It is fully adjusted 
at the factory and sharply focused upon the slit, and requires no 



FULL BIAS-NO MODULATION 




ZERO BIAS -ABOVE 
80$ MODULATION 



PARTIAL BIAS -VARYING 
WITH MODULATION BELOW &0% 

FIG. 4. Ground-noise reduction as applied to variable-width 
recording. 

subsequent adjustment although, should occasion ever arise, it could 
be adjusted in the field. Likewise the location of the optical system 
upon the recorder is determined at the factory, and permanent 
adjustment of the sound-track location assured by means of dowel 
pins. Removal of the optical system from the recorder, therefore, 
does not disturb this adjustment. Optical systems can be inter- 
changed between recorders by removing the dowel pins. 

Adjustment of the System. In practice the recordist is concerned 
with only four mechanical adjustments of the optical system. These 
are (1) lamp filament centering, which has already been discussed, 
(2) focusing the slit upon the film, (3) adjusting the galvanometer 
about the horizontal axis, and (4) about the vertical axis. The 



182 L. T. SACHTLEBEN fj. S. M. p. E. 

focusing adjustment of the slit upon the film is ordinarily fixed, 
but is easily changed and checked. The adjustment is accomplished 
by inserting a special microscope into the end of the hollow recording 
drum, and viewing the spot of light upon the film through a hole in 
the drum while moving the objective and slit assembly axially by 
means of a knurled adjusting nut. When the slit image appears 
sharp the adjustment is fixed by tightening a set-screw which clamps 
the barrel without disturbing the adjustment. 

Adjustment of the galvanometer about the vertical axis is like- 



P EU6VAT- CAMP IMTgRMEOjATe LENS 

SCREW CV.AHP ASSEMBLY CLAMP 




AOJV&T. tWti SCHEUJ 



FIG. 5. The Photophone high-fidelity, light-modulating system. 

wise made while viewing the image through the focusing microscope. 
It is turned about the vertical axis until its rotation about the hori- 
zontal axis causes the two edges of the triangle of light to reach the 
two ends of the slit simultaneously. The adjustment is then locked 
by tightening two adjusting screws which work in opposition. Ad- 
justment about the horizontal axis is associated with the adjustment 
of the ground-noise reduction system, to be described later. It is 
accomplished by turning a finely threaded screw at the rear of the 
galvanometer which, after adjustment, is secured by a knurled 
lock-nut. 

Inasmuch as the system is sealed by optical surfaces at all points, 



Aug., 1935] THE PHOTOPHONE SYSTEM 183 

dust does not enter, and it is necessary only to clean these readily 
accessible surfaces regularly with lens paper to keep the system in 
good working condition. 

The One-Fourth Mil Slit Image. The analyses of slit effect pre- 
viously mentioned show that the aperture effect diminishes rapidly 
as the slit image width decreases. Steady progress has been made 
in the direction of eliminating it. In the early days of recording, 
slit images one mil wide were considered satisfactory. Galvanom- 
eter mirrors were then too small to illuminate very much narrower 
images properly, and the limited frequency ranges then employed 
made the aperture effect not serious. A little later, three-quarter 
mil slit images came to be employed, and the way was made clear 
for further reductions by the increased efficiency of the larger mirror 
of a new galvanometer, 7 which at the same time extended the fre- 
quency range of sound-film recordings. The frequency range 4 was 
then extended still further by the design of a new galvanometer 
mechanism ; and certain increases in optical efficiency, made possible 
by the adoption of the double-edged variable-width track, were 
used to reduce the slit image width further, to one-half mil. Finally, 
the one-quarter mil recording slit image was adopted, the necessary 
increase in optical efficiency being conveniently attained by replacing 
the 5-volt, 6-ampere recorder lamp with the 10-volt, 7.5-ampere 
reproducer lamp, which more completely filled the galvanometer 
mirror with light. Improvements in the sensitivity of photographic 
materials for sound recording have also contributed to the successful 
reduction of the size of recording slit images. 

The new Photophone light-modulating system records with a 
one-quarter mil slit image, which effectively reduces the aperture 
effect to negligible proportions. Although further improvements in 
this direction may eventuate, it appears that an additional reduction 
of about 30 per cent will cause the optical system to reach the limit 
of its optical resolving power. This limit is reached when the 
primary diffraction images of the source, whose distance from the 
axis increases with diminished slit width, spread out so far as not 
to enter the lens by which the slit is imaged upon the film. An 
image can not be formed when only the central diffraction image of 
the light-source enters the lens. 

The Front Surface Galvanometer Mirror. An important improve- 
ment is the adoption of a new front-surface galvanometer mirror. 
Previously, a high-quality, back-silvered mirror was used with very 



184 



L. T. SACHTLEBEN 



[J. S. M. P. E. 



good results, except for the unavoidable secondary images char- 
acteristic of this kind of mirror. The new optical system employs 
an identical glass mirror except that the front surface is the reflecting 
surface, being coated with a film of aluminum alloy by the evapora- 
tion process. Treatment subsequent to deposition renders the 
surface resistant to corrosion and thoroughly cleanable without 
danger of mechanical injury or scratching. An additional advan- 
tage, attending the elimination of the secondary images by the 
front-surface mirror, is the elimination of the stray light that for- 
merly arose from irregular reflection at the edges of the mirror. The 
total effective reflectivity of the new aluminum mirror is practically 
the same as that of the silvered mirror. 



PRINT 



IMAGE OF RECORDING APERTURE 
ON SLIT 




FIG. 6. The variable- width, push-pull sound-track. 

Instant Convertibility from Conventional to Push-Pull Recording. 
Within recent months, the RCA engineers have developed an im- 
proved system for noiseless recording known as the push-pull system. 8 
Briefly, the central feature of the new system is a new type of sound- 
track, which is opaque except where modulation has been impressed 
upon it. This sound-track is effectively a double track, occupying 
the same portion of the film as the standard variable-width tracks, 
with one of the double tracks carrying the positive portion of the 
impressed wave, and the other alongside it and carrying the negative 
portion of the wave. In reproduction, these positive and negative 
impulses are electrically recombined so as to recreate the original 
sound. 

The essential design of the optical system undergoes no change 
when converted for the production of this new push-pull sound-track. 



Aug., 1935] THE PHOTOPHONE SYSTEM 185 

The triangular recording aperture used in recording the conventional 
variable-width sound-track is replaced by an aperture with two 
triangular openings (Fig. 6) so arranged that when modulation 
swings the galvanometer in one direction from the zero position the 
portion of the sound-track carrying the positive portion of the wave 
is exposed to a varying width, and when it swings it in the opposite 
direction the negat ve portion of the sound-track is exposed in the 
same manner. When there is no modulation the sound-track re- 
ceives no exposure, and the corresponding portions of the print are 
opaque. 

The new Photophone light-modulating system is available with 



SLIOf IN POSITION FOR 
CONVENTIONAL RECORDING 





SLIDE IN POSmOtf FOR 
PUSH PUU RECORDING 



FIG. 7. The Photophone light-modulating system equipped with 
slide for changing from conventional to push-pull variable-width re- 
cording. 

both the conventional and push-pull recording apertures mounted 
upon a single slide (Fig. 7) in such manner that simply by pushing 
the slide in or pulling it out with the fingers the optical system is 
instantly made ready to record either conventional or push-pull 
sound-tracks. When recording in conventional fashion, the ground- 
noise reduction equipment is, of course, employed and associated 
adjustments of the optical system must be made. No ground-noise 
reduction equipment is used when recording push-pull, as the push- 
pull track is inherently noiseless. 

Recording Variable-Density. Although the Photophone light- 
modulating system is designed primarily for variable-width recording, 
it lends itself very readily to variable-density work. Since the coiled 
tungsten filament of the 10-volt, 7.5-ampere lamp is set with its 



186 



L. T. SACHTLEBEN 



[J. S. M. P. E. 



axis in the horizontal plane, a horizontal edge placed between the 
lamp and the first condenser lens will cast a straight shadow upon 
the recording aperture, displaying the well-known penumbra char- 
acteristic of shadows produced by extended sources (Fig. 8). The 
shadow is, in turn, imaged upon the slit by the same lens that images 
the recording aperture for variable-width work. The illumination 
in this penumbra will at any point be proportional to the distance 
of the point from the edge of the umbra, up to the limit of full illumi- 
nation. While Fig. 8 illustrates schematically the formation 



HORIZONTAL EDGE 



LAMP 

FILMENT, 



CONDENSER 




RECORDING 
APERTURE, 

UMBRA 



PENUMBRA 



FULL 
ILLUMINATION 



FORMATION OF PENUMBRA 




PENUMBRA 




UNBIASED 



BIASED FOR LOW MODULATION 



FIG. 8. 



CHARACTER OF APERTURE IMAGE AT SLIT 

Formation of penumbra for variable-density recording 
with the Photophone light-modulating system. 



of the penumbra, actually the top and the bottom of the filament are 
masked off at the filament image on the galvanometer mirror. This 
is done to attain a linear distribution of light in the penumbra. 
Variable-density recordings are made by modulating the galvanom- 
eter in the same way as in variable-width recording, and causing 
the penumbra of the shadow to pass up and down across the slit 
to vary the exposure of the film within the permissible limits. This 
method of producing variable-density recordings combines two de- 
sirable features, namely: sufficient illumination to expose the slower, 
finer-grained emulsions of high resolving power; and a recording 
slit image width which is fixed and equal to 0.00025 inch. The 



Aug., 1935] 



THE PHOTOPHONE SYSTEM 



187 



ground-noise reduction system operates as in variable-width work, 
and biases down the average exposure 9 of the negative for low modu- 
lation values by causing the darker portions of the penumbra to be 
cast upon the slit. 

The Direct Visual Monitoring System. An outstanding feature of 
the Photophone light-modulating system is direct visual monitoring 
by observing the vibrating image of the recording aperture. What 
is observed is not the image of the recording aperture itself, but the 
image of another aperture (i in Fig. 2) which moves through the 
same angle, being formed by the same lens that images the recording 

MONITORING SCREEN AND MONITORING IMAGE 






(a) 



(b) 



(C) 






SLIT AND APERTURE IMAGE 

FIG. 9. Adjustment of the Photophone monitoring 
system. 

aperture and being deflected by the same mirror. This image is 
reflected to a white screen (k, Fig. 2) farther from the galvanometer 
mirror than the slit and vibrates at a greater linear amplitude than 
the image of the recording aperture, making it somewhat easier to 
observe than the latter. Adjustment is simple in that it is required 
only to mark upon the card the position of one end of the monitoring 
image when the tip of the triangular recording image just crosses 
the slit (a, Fig. 9), and again when the galvanometer is in such 
position as just to illuminate the full length of the slit (b, Fig. 9). 
These points are easily determined visually and marked upon the 
monitoring card, and locate the limits beyond which the sound-track 
will be overloaded or "overshot." It is necessary to draw accurately 



188 



L. T. SACHTLEBEN 



[J. S. M. P. E. 



an additional line midway between the two, and it is by this line 
that the galvanometer is set at the mid-position about which vibra- 
tion must take place when the track is fully modulated (c, Fig. 9) . 

With the indexed monitoring card as a guide, the routine adjust- 
ments consist of mechanically setting the monitoring image to the 
central line (c), increasing the d-c. biasing current of the ground- 
noise reduction system until the mirror is deflected to the point 
where the tip of the triangle just extends across the slit (a), and, by 



MONITORING SCREEN 




FIG. 10. 



Light-modulating system and monitoring screen in 
position on a Photophone recorder. 



means of the 1000-cycle oscillator, adjusting the gain of the ground- 
noise reduction amplifier until the current that biases the mirror 
cuts off at a predetermined value of modulation, usually about 
80 per cent. These adjustments make the system ready to operate. 
During recording, the amplifier gain is kept such that the edge of 
the monitoring image does not pass beyond the limit lines on the 
monitoring card. Fig. 10 shows the appearance of the light-modulat- 
ing system and monitoring screen in position on a Photophone 
recorder. 

This visual monitoring of film recordings parallels the visual ex- 
amination of lateral-cut disk records in that overshooting is at once 



Aug., 1935] 



THE PHOTOPHONE SYSTEM 



189 



apparent to the recordist when the take is made, and retakes can 
be made at once without the necessity of first processing the record 
to learn if overshooting has occurred. 

The Recording Galvanometer. The recording galvanometer or 
vibrator is both simple and rugged in construction, and is so designed 
that its operating characteristics are stable and not dependent upon 
any field adjustments. All parts are of substantial dimensions and 
the nature of the operating characteristics of the instrument depends 
more upon the physical properties of the materials used and their 
physical dimensions than upon delicate mechanical adjustments 
subsequent to assembly. This was not true of the Duddell type of 
oscillograph vibrator originally used, in which critical adjustment of 




FIG. 11. Diagram of Photophone galvanometer or vibrator. 

the tension of a pair of delicate vibrating ribbon conductors was 
necessary. 

The construction of the galvanometer has been previously de- 
scribed, and has undergone little modification, except that tungsten 
loaded rubber damping is used, and its method of application is 
modified and improved. Fig. II 4 shows the construction: A silicon 
steel armature, a, is clamped between two laminated silicon steel 
pole-pieces, b, being separated from the pole-pieces by two non- 
magnetic spacers, e. The free end of the armature is ground to form 
a knife-edge, which fits into a groove in the semicylindrical mirror 
plate, k. A phosphor bronze ribbon is fastened to two prongs, h, 
and passes over the mirror plate. The two prongs press against the 
pole pieces and tend to spring apart, providing a small tension in the 
ribbon. The slight angle between the ribbon and the face of the 
pole piece results in a component of force tending to hold the mirror 
plate against the armature. An aluminum alloy coated front- 



190 L. T. SACHTLEBEN [J. S. M. P. E. 

surface mirror, 0.125 inch long by 0.100 inch wide, is cemented to 
the mirror plate, the cement also preventing relative motion between 
the mirror plate and the ribbon. A force applied near the end of the 
armature deflects it in a manner similar to a cantilever beam. Since 
the phosphor bronze ribbon prevents lateral displacement of the 
mirror plate, the latter is free to vibrate only rotationally about a 
center through the ribbon. A major part of the controlling stiffness 
is due to the armature itself, the remainder being in the ribbon. A 
portion of the flux from two cobalt steel magnets passes through the 
two air gaps, g. Two coils, c and d, surround the armature, but 
are not in contact with it. Coil c carries the voice current from the 
recording amplifier while coil d, wound with many turns of fine wire, 
carries the biasing current required in eliminating ground-noise. 
A rubber pad, /, provides the desired damping at resonance, which 
occurs at 9000 cycles. 

The impedance of the galvanometer is 2 ohms and its power 
consumption is less than 175 milliwatts when working at full modula- 
tion. Inasmuch as magnetic saturation occurs at 100 per cent over- 
load, the instrument is self-protected against mechanical injury, 
because beyond saturation practically no increase of deflecting force 
occurs with increased current. 

Adequate Exposures with Low Lamp Currents. In the Photophone 
light-modulating system, the illumination of the slit image is high 
and also very uniform. Negatives developed to high gammas show 
no measurable variation of density across the sound-track. Gen- 
erally speaking, the finer-grained emulsions of high resolving power 
are most desirable for sound recording, and this system provides 
sufficient light to expose such emulsions properly. At the present 
time, variable-width negatives are being recorded on Eastman 
emulsion No. 1359 and Du Pont emulsion No. 201, both of which 
are faster emulsions than the Eastman No. 1301 used in the early 
days of sound-film recording. With these emulsions and the one- 
quarter mil slit image, suitable negative densities result from operat- 
ing the 7.5-ampere lamp at 7.0 amperes. The resulting extension of 
lamp life, uniform lamp performance, and freedom from untimely 
burn-outs and frequent lamp replacements are obvious. 

Special acknowledgment is due Mr. G. L. Dimmick who, for the 
past five years, has been immediately responsible for the development 
and continuous improvement of the Photophone light-modulating 
system. 



Aug., 1935] THE PHOTOPHONE SYSTEM 191 

REFERENCES 

1 COOK, E. D.: "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV (June, 
1930), No. 6, p. 650. 

2 FOSTER, D.: "The Effect of Exposure and Development on the Quality of 
Variable- Width Photographic Sound Recording," J. Soc. Mot. Pict. Eng., XVII 
(Nov., 1931), No. 5, p. 749. 

3 HARDY, A. C.: "The Optics of Sound Recording Systems," Trans. Soc. 
Mot. Pict. Eng., XII (1928), No. 35, p. 760. 

4 DIMMICK, G. L., AND BELAR, H.: "Extension of the Frequency Range of 
Film Recording and Reproduction," /. Soc. Mot. Pict. Eng., XIX (Nov., 1932), 
No. 5, p. 401. 

6 MEES, C. E. K.: "Some Photographic Aspects of Sound Recording," /. 
Soc. Mot. Pict. Eng., XXIV (April, 1935), No. 4, p. 285. 

6 READ, S., JR.: "RCA Victor High-Fidelity Film Recording Equipment," 
J. Soc. Mot. Pict. Eng., XX (May, 1933), No. 5, p. 396. 

7 DIMMICK, G. L.: "Galvanometers for Variable- Area Recording," /. Soc. 
Mot. Pict. Eng., XV (Oct., 1930), No. 4, p. 428. 

8 DIMMICK, G. L., AND BELAR, H.: "An Improved System for Noiseless Re- 
cording," /. Soc. Mot. Pict. Eng., XXIII (July, 1934), No. 1, p. 48. 

9 SILENT, H. C., AND FRAYNE, J. G.: "Western Electric Noiseless Recording," 
J. Soc. Mot. Pict. Eng., XVIII (May, 1932), No. 5, p. 551. 

DISCUSSION 

MR. KELLOGG: Several improvements have been mentioned that have re- 
sulted in increased light intensity. Advantage has been taken of this to narrow 
the recording slit and thereby improve the resolution. The most important 
factor in increasing the light intensity was undoubtedly the development by 
Dimmick of a galvanometer having a much larger mirror than the oscillograph 
type of galvanometer previously employed. Next in importance was the feature 
of placing the mirror axis horizontal and using an acute angle of intersection 
between the slit and the edge of the aperture image, thereby making it possible 
to cover the complete range of slit illumination with smaller movements and with 
the galvanometer brought closer to the slit. A third major factor in making 
it feasible to reduce the slit to the present quarter-mil width was the availability 
of faster recording films. This contribution through the cooperation of film 
manufacturers deserves special acknowledgment. The problem was evidently 
by no means a simple one. Motion picture positive which was used at first is 
a fine-grained, high-resolution film. Increased speed, although desirable, has 
not been extremely urgent. At first it seemed as if any increase in speed was 
at the cost of resolution; and high contrast, which is desirable for variable- 
width recording, was always accompanied by a higher toe. But films were 
finally made available which had the increased speed with little, if any, sacrifice 
of the other desirable properties. 



REPORT OF THE STANDARDS COMMITTEE* 

With the publication of the new motion picture standards in the 
November, 1934, issue of the JOURNAL, and of the Standards Booklet, 
the Standards Committee, under the Chairmanship of M. C. Batsel, 
completed many of the problems in which it had been engaged. 

At the beginning of 1935, L. A. Jones, Engineering Vice-President, 
appointed a new Standards Committee, composed of many of the old 
members and some new members. One special feature of this new 
Committee is that there is a Vice-Ch airman on the Pacific Coast 
whose duties are to help correlate the opinions of the West Coast 
members with those of the East Coast members. J. A. Dubray was 
appointed to this Vice- Chairmanship. Before each meeting of the 
Standards Committee, an agenda is mailed to all the members, and 
those on the West Coast meet together at about the same time as 
those on the East Coast, and discuss the same problems. The min- 
utes of both groups are sent to all the members. 

The main items under consideration by the Committee at the 
present time are as follows: 

(1) Sprockets. The standards published in the Standards Booklet 
for 35-mm. sprockets apply only to projectors. We have no stand- 
ards for cameras, sound apparatus, or printers. There is considerable 
doubt in the minds of most members of the Committee as to whether 
or not we are ready to standardize any of these items, or even whether 
it will ever be desirable to do so. The subject is being taken up with 
the manufacturers of these products to see what, if anything, should 
be done at the present time. 

(2) Guiding Methods in Cameras, Printers, Etc. The Standards 
Committee has already committed itself to edge guiding in the pro- 
jector, but as yet no agreement has been reached as to the desirability 
of edge guiding in the camera or printer. 

(3) Screen Brightness. The Standards Committee is awaiting the 
report of the Committee on Projection Screen Brightness before tak- 
ing any action upon the subject of standard screen brightness. 

(4) Reel Dimensions. The question of standardizing sizes and di- 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
192 



REPORT OF STANDARDS COMMITTEE 193 

mensions of reels was ably discussed in the report of the Sub-Com- 
mittee on Exchange Practice at the Spring Meeting in 1934, and is 
now under consideration by the Standards Committee. 

(5) Photoelectric Cells. A set of standard specifications for photo- 
electric cells adopted by the British Standards Institution and pub- 
lished in January, 1935, is under consideration and is awaiting criti- 
cism by the Sound Committee. 

Two of the standards that have already been officially adopted by 
the Society are still of great interest. One of these relates to the use 
of a single type of perforation for both negative and positive, and it is 
the hope of the Committee that the old type of perforation will be 
discarded for negative film as it has been for positive film. The other 
item, the standards for 16-mm. sound-film, is the source of consider- 
able anxiety to those interested in the future of 16-mm. projection. 

Through a misinterpretation of the drawing published in the 
JOURNAL of the Society of Motion Picture Engineers, the Committee 
of the Deutsche Kinotechnische Gesellschaft recommended a film 
which they thought was a duplicate of ours but which actually had 
the sound-track upon the opposite side of the film from that of the 
American standard. Through a misinterpretation of their drawing, 
upon our part, this error was not noticed until some time had passed, 
with the result that two different camps have arisen, each favoring 
a different standard. 

Last summer, at a conference called by the International Educa- 
tional Cinematographic Institute (I. C. E.) of the League of 
Nations, the German proposal was adopted. All the above was ex- 
plained in more detail by Mr. Batsel in the report of the Standards 
Committee for last Fall. Since then, a Sectional Committee on 
Motion Pictures, organized according to the procedure of the Ameri- 
can Standards Association, has been formed. This Committee will 
undoubtedly approve the S. M. P. E. standards, and it is hoped that 
the American Standards Association, through the International 
Standards Association, can succeed this next summer in achieving in- 
ternational adoption of a single standard, through regular channels. 

During the last week in April, 1935, an International Film Congress 
was held in Berlin and a sub-committee of this Congress was formed 
for further consideration of the 16-mm. sound-film standard. Al 
though the Standards Committee has not yet seen the full report of 
this meeting, it apparently was in favor of upholding the decision of 



194 REPORT OF STANDARDS COMMITTEE 

the Stresa Conference. The subject will again be discussed at a con- 
ference of the International Standards Association to be held at Paris 
in July in conjunction with the International Photographic Congress, 
and the results of this conference will be of great interest to the in- 
dustry. 

Aside from these items, there are a few items upon which the Com- 
mittee has not yet had a chance to act. These are : 

(1) A proposal for 8-mm. sound-film. 

(2) The possibility of supplying sets of densities of sputtered glass or other 
material for use as reference standards with densitometers. 

(3) The production of a 16-mm. test-film for use in testing 16-mm. sound 
projectors. 

The general policy of the Standards Committee remains unchanged 
except that we shall endeavor to keep in closer touch with the Euro- 
pean standardization groups and to promote a closer cooperation be- 
tween the East Coast and West Coast members. 

E. K. CARVER, Chairman 
J. A. DUBRAY, Vice-Chairman 

M. C. BATSEL R. C. HUBBARD H. RUBIN 

W. H. CARSON E. HUSE O. SANDVIK 

A. C. DOWNES C. L. LOOTENS H. B. SANTEE 

P. H. EVANS K. F. MORGAN J. L. SPENCE 

R. E. FARNHAM N. F. OAKLEY H. M. STOLLER 

C. L. FARRAND G. F. RACKETT R. C. WILLMAN 

H. GRIFFIN W. B. RAYTON A. WISE 

C. N. REIFSTECK 



FALL, 1935, CONVENTION 



WASHINGTON, D. C. 

WARDMAN PARK HOTEL 

OCTOBER 21-24, INCLUSIVE 



Officers and Committees in Charge 

PROGRAM AND FACILITIES 

W. C. KUNZMANN, Convention V ice-President 

J. I. CRABTREE, Editorial Vice-President 

J. O. BAKER, Chairman, Papers Committee 

LOCAL ARRANGEMENTS AND RECEPTION COMMITTEE 

NAT GLASSER, Chairman 

CAPT. JOHN G. BRADLEY A. DE TITTA N. C. HAEFELE 

J. T. CORRIGAN H. C. BUSMAN J. R. MCLEMORE 

H. T. COWLING R. EVANS F. J. STORTY 

N. D. GOLDEN 

PROJECTION COMMITTEE 

H. GRIFFIN, Chairman 

J. FRANK, JR. J. R. MCLEMORE 

N. C. HAEFELE A. L. RAVEN 

F. J. STORTY 

APPARATUS EXHIBIT 

O. F. NEU, Chairman 
J. FRANK, JR. S. HARRIS 

BANQUET 

W. C. KUNZMANN, Chairman 

CAPT. JOHN G. BRADLEY A. DE TITTA N. D. GOLDEN 

H. T. COWLING R. EVANS S. HARRIS 

PUBLICITY 

W. WHITMORE, Chairman 

J. R. CAMERON G. E. MATTHEWS 

J. J. FINN P. A. McGuiRE 

F. H. RICHARDSON 

195 



196 FALL, 1935, CONVENTION [J. S. M. P. E. 

MEMBERSHIP 

O. M. GLUNT, Financial Vice-P resident 

E. R. GEIB, Chairman, Membership Committee 

LADIES' RECEPTION COMMITTEE 

MRS. H. T. COWLING, Hostess 

assisted by 

Miss A. C. BECKER Miss EVELYN GLASSER MRS. N. D. GOLDEN 

MRS. JOHN G. BRADLEY MRS. NAT GLASSER MRS. N. C. HAEFELE 
MRS. R. EVANS Miss H. O'CONNOR 

Headquarters 

The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations and Convention facilities are assured. Registration 
will begin at 9 A.M., Monday, October 21st. A special suite will be provided for 
the ladies. Rates for S. M. P. E. delegates, European plan, will be as follows: 

One person, room and bath $3 . 00 

Two persons, double bed and bath 5. 00 

Two persons, twin beds and bath 5 . 00 

Rates for connecting parlors 5 . 00 

A modern fire-proof garage is located on the Hotel property, and a special 75 cents 
per day rate has been arranged for. 

Technical Sessions 

An attractive program of technical papers and presentations is being arranged 
by the Papers Committee. Sessions will be held in the Little Theater of the Hotel, 
off the west lobby, as follows: Monday to Thursday mornings, inclusive; and 
Monday, Tuesday, and Thursday afternoons. 

Film Programs 

Exhibitions of newly released motion picture features and short subjects will be 
held in the Little Theater on Monday and Tuesday evenings. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held in the east 
lobby of the Hotel, to which all manufacturers of equipment are invited to con- 
tribute. The apparatus to be exhibited must either be new or contain new fea- 
tures of interest from a technical point of view. Information concerning the 
exhibit and reservations for space should be made in writing to the Chairman of 
the Exhibits Committee, Mr. O. F. Neu, addressed to the General Office of the 
Society at the Hotel Pennsylvania, New York, N. Y. No charge will be made for 
space. 



Aug., 1935] FALL, 1935, CONVENTION 197 

Semi- Annual Banquet 

The semi-annual banquet and dance of the Society will be held in the Continental 
Room of the Hotel on Wednesday, October 23rd. Addresses will be delivered 
by eminent members of the industry, followed by dancing and entertainment. 
Tables reserved at the registration desk. 

Recreation 

The Wardman Park Hotel management is arranging for golfing privileges for 
S. M. P. E. delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 

PROGRAM 

Monday, Oct. 21st 

9 : 30 A.M. Registration 
10 : 00 A.M. Society business 

Technical papers program 
12 : 30 P.M. Informal get-together luncheon 

2: 00 P.M. Technical papers program 

8 : 00 P.M. Exhibition of newly released motion pictures 

Tuesday, Oct. 22nd 

10 : 00 A.M. Technical papers program 
2 : 00 P.M. Technical papers program 
8 : 00 P.M. Exhibition of newly released motion pictures 

Wednesday, Oct. 23rd 

10 : 00 A.M. Technical papers program 

12 : 30 P.M. Free afternoon, for recreation or special trips and visits 
7:30 P.M. Semi-annual banquet 

Thursday, Oct. 24th 

10 : 00 A.M. Technical papers program 
2 : 00 P.M. Technical papers program 
6:00 P.M. Adjournment of the Convention 



SOCIETY ANNOUNCEMENTS 

BOARD OF GOVERNORS 

At a meeting held at the Hotel Pennsylvania, New York, N. Y., October 19th, 
further plans were made for the approaching Convention at Washington, D. C., 
October 21st-24th, inclusive. The general arrangements for the Convention are 
described in the preceding pages of this issue of the JOURNAL, and preliminary 
programs and Hotel reservation cards will be mailed to the membership in the 
near future. 

Nominations were held for officers and governors of the Society for 1936. 
Voting ballots will be mailed shortly to the Fellow and Active membership of 
the Society, the ballots will be counted and the results announced at the Wash- 
ington Convention, and the successful candidates will assume their offices on 
January 1, 1936. 

SECTIONAL COMMITTEE 

Following the Reichsfilmkammer (National Film Conference) held at Berlin 
last April, at which the differences existing between the DIN (Deutschen Indus- 
trie Normen) standards and the SMPE standards for 16-mm. sound-film were 
discussed, arrangements were made to confer upon the subject further at Paris 
during the week of July 7th, under the auspices of the International Standards 
Association and the International Congress of Scientific and Applied Photography. 
An extensive presentation of the American arguments was drawn up, and Mr. 
George Friedl, who represented the Sectional Committee at the Berlin confer- 
ence, was appointed American delegate to the Paris Congress. Mr. W. J. McNair, 
assistant to Dr. P. G. Agnew, Secretary of the American Standards Association, 
also attended the Congress in the interests of the Sectional Committee. Dr. 
Walter Clark, Secretary of the American Committee of the International Con- 
gress, attended in his official capacity. A report of the proceedings of the Congress 
will be published in the JOURNAL as soon as available. 



HOLLYWOOD CONVENTION PHOTOGRAPHS 

Those who were in the photograph taken at the Fox Hill Studio on June 25th 
and who have not yet received a copy, may do so by writing to the General 
Office of the Society, enclosing six cents in stamps as return postage. No charge 
will be made for the photograph. 

At the Warner Bros. -First National Studios, on June 26th, two photographs 
were taken, the first in two halves, which may be matched and glued together. 
Although the match is not exact, the picture is quite good. Copies may be 
obtained from the General Office at the price of fifty cents for the two sections, 
including postage. The other photograph is identical to the larger one, but of 
198 



SOCIETY ANNOUNCEMENTS 199 

half the width. The smaller picture measures 8 by 10 inches, and the larger 
8 by 20. 

ACADEMY OF MOTION PICTURE ARTS AND SCIENCES 

Ralph Townsend, Assistant Director of Sound Recording for Fox studios, 
was elected Chairman of the Research Council Acoustic Subcommittee recently 
at the first meeting of the group. 

In addition to the acoustic tests now being planned, the Subcommittee also 
decided to make a questionnaire survey to determine the type of wall coverings, 
floor coverings, and set materials used in each studio, and to include a correlation 
of this information in their report, which will be submitted to the Research 
Council for distribution to the studio sound, art, and construction departments. 

SOCIETY SUPPLIES 

Reprints of Standards of the SMPE and Recommended Practice may be obtained 
from the General Office of the Society at the price of twenty-five cents each. 

Copies of Aims and Accomplishments, an index of the Transactions from Octo- 
ber, 1916, to June, 1930, containing summaries of all the articles, and author and 
classified indexes, may be obtained from the General Office at the price of one 
dollar each. Only a limited number of copies remains. 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 
JOURNAL, may be obtained from the General office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



STANDARD S. M. P. E. 

VISUAL AND SOUND TEST REELS 

Prepared under the Supervision 

OF THE 
PROJECTION PRACTICE COMMITTEE 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 
Reels sold complete only (no short sections). 



PRICE $37.50 FOR EACH SECTION, 
INCLUDING INSTRUCTIONS 



(Shipped to any point in the United States) 

Address the 
SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL PENNSYLVANIA 
NEW YORK, N. Y. 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV SEPTEMBER, 1935 Number 3 



CONTENTS 

Page 

A Comparison of Variable-Density and Variable- Width Sys- 
tems E. W. KELLOGG 203 

Relation Between Illumination and Screen Size for Non- 
Theatrical Projection D. F. LYMAN 227 

Sensitometric Studies of Processing Conditions for Motion 
Picture Films H. MEYER 239 

New Emulsions for Special Fields in Motion Picture Photog- 
raphy W. LEAHY 248 

Technical Aspects of the Motion Picture. .A. N. GOLDSMITH 254 

Some Technical Aspects of Recording Music. R. H. TOWNSEND 259 

Report of the Projection Screen Brightness Committee 269 

Apparatus Symposium: 

Three New Kodascopes N. B. GREEN 271 

A New Sound Reader and Frame Viewer 

I . SERRURIER 275 

Arc Supply Generator for Use with Suprex Carbons 

W. K. HARTMAN 278 

A Professional 16-Mm. Projector with Intermittent Sprocket 

H. A. DEVRY 279 

Eugene Augustin Lauste 281 

Fall, 1935, Convention at Washington, D. C 284 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTRBE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



A COMPARISON OF VARIABLE-DENSITY AND 
VARIABLE-WIDTH SYSTEMS* 

E. W. KELLOGG** 

Summary. In the earlier stages of the development of the sound recording system 
of RCA Photophone, the balance seemed to swing at times in favor of variable-density 
and at times in favor of variable-width recording. The close tolerances for exposure 
and development which the variable-density system requires made the variable-width 
system seem preferable. Difficulties in obtaining, with narrow recording slits, 
sufficient exposure for variable-width recording on high-resolution films were over- 
come by various improvements in the optical recording system and galvanometers. 
A slit image 0.00025 inch wide is standard for the newest recording equipment. 

While for variable-density recording a number of combinations of exposure and 
development of both negative and positive have been worked out which give a linear 
relation between exposing light and print transmission, practical considerations, 
especially that of obtaining reasonably high outputs, have led to departures from the 
ideal conditions. Measurements have been published which show the wave-form 
distortion found in actual recordings of both types, made under representative condi- 
tions. The measurements indicate considerably greater distortion in the variable- 
density records for amplitudes in excess of 50 per cent modulation. For smaller 
modulations the distortion in both systems is of the order of 5 per cent or less . Variable- 
width records are subject to a wave-shape distortion which becomes appreciable at 
very high frequencies. Printing to a suitable density largely neutralizes this type of 
distortion. 

The output from variable-width records is higher by 3 to 8 db. With respect to 
ground-noise due to scratches and dirt on the positive, the variable-density system 
may be expected to show a better ratio; while the hiss due to film graininess, and noise 
due to imperfections of the negative, is less in variable-width records. The fact that 
extension of the frequency range makes the latter type of noise a more serious factor, 
plus the fact that the prints will be relatively new, puts the variable-width system at 
advantage for first-run theaters and for master records for re-recording. 

The system of monitoring by observing the moving light spot in variable-width 
recorders and the ability to judge a recording by inspection of the negative are of 
considerable practical importance. The full benefits of the recently developed "push- 
pull" sound-track are obtained only with variable-width recordings. 

HISTORICAL BACKGROUND 

Development of the system of recording sound on film as now 
employed by RCA Photophone began with the early work of C. A. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RCA Manufacturing Company, Camden, N. J. 

203 



204 E. W. KELLOGG [J. s. M. p. E. 

Hoxie at Schenectady. As with many such developments, the early 
experiments were primarily concerned with the general problem of 
recording the sound, and immediate concern was not given to specific 
applications. Although employing motion picture film, Hoxie used 
as much of the film as seemed desirable at the time, and did not try 
to leave room on the same film for the picture. Many of his records 
utilized the entire one inch of clear space between the perforations. 
In view of the increasing difficulty of obtaining sufficient exposure 
as the sound-track is widened, it is somewhat difficult to explain 
how Mr. Hoxie got sufficient exposure, except that he used negative 
stock, a fairly wide recording slit, moderately low film speed (60 
feet per minute), and probably an extra large mirror on the galvanom- 
eter. Extreme care of slit alignment was, of course, necessary. 
With many of these records Hoxie obtained a very creditable quality 
of sound reproduction. The recording of speech was the primary 
interest. The work was done in the same laboratory where the 
General Electric oscillographs were developed and built, and oscillo- 
graph tradition naturally exercised some influence on point of 
view, so that it was natural to think of a sound-track in terms of the 
rather large dimensions of conventional oscillograms. The recording 
optical system was, in principle, essentially the same as that em- 
ployed in an oscillograph and produced a variable-width track. 
All who have worked with oscillographs are aware of the fact that 
it is not too easy to obtain a sharp spot. In the recording 
arrangement the edge of the light-spot was not sharply defined. 
In fact the light intensity tapered off for a considerable distance. 
We debated the pros and cons of the "fuzzy edge." If no over- 
shooting is permitted, the space across the sound-track which the 
tapering light intensity occupies is track width thrown away, so 
far as possible light modulation is concerned. On the other hand, 
if overshooting does occur the distortion is less noticeable and less 
objectionable with the soft-edged light spot. The more orthodox 
members ol the group maintained that overshooting must be strictly 
avoided ; but Mr. Hoxie, who was of a very practical turn of mind, was 
not convinced that that was important. To prove his point he 
masked off the ends of his reproducing slit and played one of his 
records using various fractions of the total width. To his surprise 
and to the surprise of the other listeners no sudden ruin of quality 
occurred, and the record still sounded good as the scanning slit was 
shortened, down to Vie inch. The realization that such a narrow 



Sept., 1935] COMPARISON OF SYSTEMS 205 

track was entirely practicable led to the happy conclusion that picture 
and sound could just as well as not be put on one film, and from then 
on only narrow tracks were employed. 

Obviously, the narrow strip of track used in Hoxie's test could 
reproduce sound only by its variations in density; and, as he shortened 
his scanning slit, he had gone by steps from a variable-width to a 
variable-density track. For a while the variable-density track 
reigned supreme at Schenectady. These variable-density tracks 
were made by providing graded light intensity in the direction of 
slit length and sweeping the light-spot back and forth through 
amplitudes several times the slit length. The method was clumsy, 
and expensive of power for vibrator input, but was nevertheless 
effective. 

The revival of interest in the variable-width system is probably 
due to A. C. Hardy, who called our attention to the much greater 
danger of distortion in the variable-density system due to lack of 
linearity in the film characteristics, 1 unless exposures and develop- 
ments were held within very close tolerances. To make a true 
variable-width track only 1 /i inch wide or less called for a new 
optical system capable of producing at the film a light-spot whose 
edge was much sharper than an optical system such as those used 
in oscillographs was capable of giving. The desired result was 
obtained by the simple expedient of forming on the film a reduced 
image of a comparatively long slit over which the light-spot played 
as before. In this small image, the edge of the light-spot was sharp- 
ened in the same ratio that the slit was shortened. Although there 
have been a number of modifications of the optical system, this 
general feature has been retained. 

A desire for achieving the best possible results in sound quality 
led to experiments with various film emulsions and different re- 
cording systems. Many engineers concerned with sound recording 
and reproduction have a feeling that devices without moving parts 
are likely to be inherently superior, especially because adequate 
reproduction of the high frequencies has always been a problem, 
and avoidance of limitations due to inertia seemed like a step in the 
right direction. This idea led to studies of the possibilities of glow- 
lamp recording and particularly to experiments with Kerr cells, 
both at the General Electric Company and the Westinghouse Com- 
pany. 2 These, of course, could make only variable-density tracks. 
While there is something to be said for the avoidance of moving 



206 E. W. KELLOGG [J. S. M. p. E. 

mechanical parts, the oscillograph galvanometers were, as subsequent 
experience has proved, one of the least of the sources of trouble and 
distortion. 

The spread of exposure in the film was early appreciated as one 
of the major causes of distortion and loss of high-frequency response. 
How much of these losses to attribute to the film and how much to 
imperfections in the optical systems was difficult to decide, and 
still is; but the obvious and proper procedure was to improve both 
as much as possible. Films of higher resolving power called for 
greater exposure, and this factor for a while turned the balance in 
favor of variable-density. Those who have analyzed and calculated 
light intensity in optical systems are familiar with the fact that 
high light intensities are easier to obtain in a variable-density system. 
The Kerr cell was certainly not a help in that direction, owing to the 
necessity of passing the light between closely spaced plates through 
a considerable thickness of the yellow nitrobenzol. On the other 
hand, a single-string light-valve 3 was built and recordings made 
with it which were regarded as quite satisfactory. At Prof. Hardy's 
suggestion, considerable work was done with yellow-dyed duplicating 
film, which showed especially fine grain and high resolution. 
This film was readily given adequate exposure with the light- valve, 
but meantime progress was being made with systems of the oscillo- 
graphic type adapted to variable-width records. Mr. L. E. Clark sug- 
gested the use of a cylindrical lens close to the slit, and this, although 
reducing the depth of focus, gave a marked increase in light intensity, 
so that an exposure entirely adequate for Eastman positive stock 
was obtained with a slit image about 3 /4 mil wide at the film. The 
optical system in this form was employed in Photophone recording 
equipments until 1932. 

A modification of the optical system which gave a large increase 
in light intensity without sacrificing depth of focus consisted in 
placing the galvanometer with its axis horizontal so that the light 
spot moved up and down across the slit instead of parallel with it. 4 
The light-spot was given a slanting edge, intersecting the slit at an 
acute angle. This made it possible to illuminate varying fractions 
of the slit length with much smaller movements of the light-spot than 
had previously been necessary when the motion of the spot was 
parallel to the slit. With a smaller required light-spot movement, 
it became feasible to bring the galvanometer closer to the slit. Other 
factors being kept the same, the light intensity depends on the solid 



Sept., 1935] COMPARISON OF SYSTEMS 207 

angle subtended by the mirror as viewed from the slit, and this 
solid angle increases with the inverse square of the distance between 
the mirror and the slit. With this "scissors type" optical system 
it was possible to get adequate exposures for variable-width tracks 
on yellow-dyed film. The argument for variable-density tracks on 
the score of greater ease of obtaining plenty of exposure thereafter 
had little weight. More recently the development of the magnetic 
vane galvanometer described by G. L. Dimmick, 5 with its large 
mirror, has made possible a several-fold increase in light intensity. 
Advantage has been taken of this, and of the increased speed of the 
newer recording films, to reduce the width of the recording slit, and 
a slit image J /4 m il wide at the film is standard with the newest 
Photophone equipment. The large mirror also results in utilizing 
enough of the aperture of the objective lens to avoid diffraction 
losses,* while at the same time maintaining a desirable depth of 
focus, so that focusing is not too critical. With a recording slit of 
the size and quality now obtainable, such spreading of exposures 
as occurs is due almost entirely to scattering of light within the 
film emulsion itself. 

Two arguments for variable-density tracks which for a time carried 
some weight in the minds of the development engineers were, first, 
that the variable-width system would call for a higher degree of 
light uniformity across the reproducing slit; and, second, that 
scratches in the film (of which we are today happily getting far 
fewer than in the early days of talking pictures) would cut minute 
pieces out of the wave-form and thereby produce distortion. With 
a variable-density system, on the other hand, a streak parallel to 
the track merely narrows the track by that much. Satisfactorily 
uniform slit images were obtained with the reproducing optical 
system adopted, and there appeared to be no reason to anticipate 
difficulty on this score, especially since it can be shown that fairly 
large departures from uniformity of slit illumination result in sur- 
prisingly little wave-shape distortion.** To ascertain whether the 

* In the recording optical system, an image of the mirror is formed approxi- 
mately in the plane of the objective lens. Since the image does not completely 
fill the lens aperture, only the central part of the lens is used. The resolving 
power of a well corrected lens increases with its numerical aperture. Utilizing 
a larger part of the available lens area results in an increase in the effective 
numerical aperture. 

** The curve which shows the relation between galvanometer deflection and 
photo-cell illumination (and which should be a straight line for zero distortion) 



208 E. W. KELLOGG [J. S. M. p. E 

wave-form distortion which was feared from parallel streaks on the 
film would be serious, tests were made by sandpapering the sound- 
track in such a direction as to make parallel scratches. In running 
the film repeatedly with more and more scratches, the only change 
which was detected was the inevitable increase in ground-noise, 
which would also be true of a variable-density track. The results 
of these tests removed the last argument in the minds of the engineers 
at Schenectady against adoption of the variable-width system. 
Continued improvements in equipment, and the experiences of the 
past few years, have in our estimation vindicated the choice. 

COMPARISON OF VARIABLE-DENSITY AND VARIABLE-WIDTH SYSTEMS 

The fact that engineers not associated with RCA chose and are 
still using the variable-density system makes it appropriate to ex- 
amine the arguments on both sides as they appear today. It would, 
of course, require a very clear case of superiority of one type over 
the other to induce any large organization to change the system 
which it has been employing. It is of interest, however, that the 
latest equipment offered by RCA Photophone happens to lend itself, 
with an extremely simple modification, to the making of variable- 
density as well as variable- width tracks. This is an almost acci- 
dental feature of the optical system in the form in which it has been 
developed, and although we are firmly convinced of the practical 
superiority of the variable-width system, this feature of our equip- 
ment is advantageous in that those who have a preference for variable - 
density tracks (due perhaps to greater familiarity with their require- 
ments), or those who wish to be on the right side and feel that the 

is the integral of the curve which shows the light intensity from point to point 
along the slit. If various, somewhat irregular, or non-uniform curves of dis- 
tribution are assumed and the corresponding total light or integral curves cal- 
culated, the latter will be found to deviate only slightly from straight lines. As 
an example of what would be a particularly bad departure from uniformity, it 
was assumed that the light intensity is down 25 per cent at each end of the slit, 
the distribution being represented by a curve of parabolic form. Calculation of 
the distortion of a sine wave showed that a third harmonic (no others) would be 
produced, having an amplitude equal to 2 per cent of the fundamental. If, 
instead of a simple sine wave track, we assume a symmetrical or double-edged 
track, such as is now employed, the same irregularity of illumination results in 
the production of a 3 per cent second harmonic and a negligible third. If the 
irregularities are of the same magnitude but more numerous, such, for example, 
as might be produced by imaging a coil filament in the slit, the harmonics are of 
higher order but are still smaller in magnitude. 



Sept., 1935] COMPARISON OF SYSTEMS 209 

question of relative merits is still in the balance, would with this 
equipment be able to make records of either type by a simple 
change. 6 

WAVE-SHAPE DISTORTION IN VARIABLE-DENSITY RECORDS 

The primary and generally recognized argument for the variable- 
width sound-track is that virtual freedom from distortion is not 
dependent on observation of close tolerances in respect to exposure 
and development of both negatives and prints. The edge of the 
light-spot moves back and forth along the length of the slit and 
traces a microscopic oscillogram on the film, whose outline is a picture 
of the wave-shape. So long as the film on one side of this edge is 
of uniform darkness and the film on the other side uniformly clear, 
we have a practically perfect trace of the sound wave, for it has been 
possible to develop a galvanometer having a response range up to 
9000 cycles and in which distortion is negligible, so far as we can 
ascertain. Only in so far as the recording slit must have a finite 
width, and as the film fails to resolve the high-frequency waves, does 
wave-shape distortion enter in the recording. Distortion, therefore, 
becomes appreciable only when this spreading of exposure covers 
a substantial fraction of a wavelength, or, in other words, at high 
frequency. 

On the other hand, as pointed out by Hardy 1 and others, 7 a truly 
linear relation between negative illumination and print transmission, 
such as is required for faithful reproduction, is obtained in variable- 
density tracks only when the exposures of both negative and print 
are within the limits of the straight parts of the H&D curves, and 
the product of the negative and print gammas is equal to unity. 
In this the gammas to be considered are not those obtained with 
sensitometers and densitometers of the forms in general use, but 
the exposures; and methods of measuring density should conform 
to the conditions of recording, printing, and reproducing in order 
that departures from photographic reciprocity and differences in 
diffusion of transmitted light may not introduce errors. Routine 
measurements may be made with standard apparatus and suitable 
correction factors applied, the correction factors depending on the 
recording, printing, and reproducing systems employed. Thus, in 
laboratories where accurate sensitometric control is attempted, it 
is general practice to check development and to measure and specify 
densities in the usual manner, but to call for diffuse densities that 



210 E. W. KELLOGG [J. s. M. p. E. 

correspond to the desired semi-specular densities and to work for 
a gamma product (in terms of diffuse densities) other than unity, 
the necessary correction factors having been taken into ac- 
count. The problem of sound-track density and gamma determi- 
nations has been treated by Tuttle and McFarlane 8 and by Mac- 
Kenzie. 9 For the relation between the ordinarily measured diffuse 
density, and the semi-specular density which is appropriate in 
rating sound prints in average reproducing optical systems, the 
authors just mentioned give factors of the order ot 1.25 to 1.3. 

If the only difficulty were that of laboratory control, in holding the 
gamma product within reasonable limits (assuming the proper 
product to have been correctly ascertained), the problem would not 
be serious, for calculation shows that only about 5 per cent harmonics 
would be produced by departures of 0.2 in either direction from the 
unity product, provided the exposures are confined to the straight 
parts of the characteristics. It appears, however, that the difficulties 
are not with maintaining the gamma product within these limits, 
but rather that it is practically necessary to utilize ranges of ex- 
posure much below the straight-line portion of the H&D character- 
istic. For example, positive film developed to a gamma of approxi- 
mately 2 is practically standard for the final print of the sound-track, 
for the reason that the picture requires this treatment. The H&D 
curve for positive film developed to a gamma of 2 does not become 
straight until the diffuse density exceeds 0.6 or more, although the 
curvature is slight if the density does not fall below 0.5. The semi- 
specular density which corresponds to a diffuse density of 0.5, 
would be about 0.63, or the transmission to the photo-cell 23 per cent. 
If the minimum transmission can be as small as 2 per cent, this leaves 
only a 21 per cent range from minimum to maximum transmission, 
which would result in very small photo-cell output. The average 
or unmodulated transmission would have to be as low as about 
12.5 per cent. In order to obtain more output, higher average 
transmissions are used, ranging from 15 to 30 per cent. At full 
modulation the maximum transmission would be nearly twice the 
average, and this would mean minimum diffuse densities of 0.42 to 
0.18, which would be well down on the toe of the H&D curve. 

There appears to be no theoretical reason why the negative might 
not be given sufficient exposure to avoid using the toe of its H&D 
characteristic, but there is no object in confining the negative ex- 
posure to the straight part of the curve unless the prints also are 



Sept., 1935] COMPARISON OF SYSTEMS 211 

to be so confined. In fact, the negative toe can be used partly to 
offset the distortion resulting from the print toe. There is, more- 
over, for variable-density recordings, as will be presently explained, 
an objection on the score of ground-noise to the use of a negative 
any darker than necessary. Instead of adhering to the classical 
means of obtaining distortionless reproduction, therefore, it has 
become general practice to utilize part of the toe of the positive 
and, in some cases, that of the negative as well, and to attempt to 
work out a combination which gives acceptably low distortion. 

A very thorough analysis of toe recording, and a comparison of 
toe recording with classical recording and with composite recording, 
was published in 1932 by MacKenzie, 9 who shows that it is possible 
to obtain an over-all characteristic having a usable linear range by 
proper combinations of toe recording, toe printing, and development. 
The toe recording, while having relatively high output, is of very 
limited range, and does not give good ground-noise ratio nor permit 
much benefit from biasing the light- valve. Strict adherence to the 
classical conditions for freedom from distortion is shown to result 
in good ground-noise ratios but extremely low output. A more 
practical solution is a composite system in which the part of the 
negative characteristic that is utilized is for the most part above the 
toe; and in which, in view of the reduced average slope of the part 
of the positive H&D characteristic which is used, comprising part of 
the toe, an effective gamma product of about 1.4 is used. The 
permissible limits of negative exposure are different for different 
printer settings. 

In his recent review of sound recording Mees 10 assumes a 25 
per cent print transmission for the unmodulated track without light- 
valve bias as fairly representative of practice in variable-density 
tracks. This is near the upper end of the toe of Eastman positive 
film developed to a gamma of 2, so that the modulation is partly 
on the toe and partly on the straight part of the H&D curve. Mees 
presents a number of over-all characteristics for various conditions, 
from which an idea can be formed of the effects of changing various 
factors. 

Since the avoidance of distortion in variable-density systems is 
largely on an empirical basis and involves certain compromises, 
no fundamental rules can be laid down to define correct procedure, 
and opinions and practices vary. The lack of universality of rules 
and specifications, of course, makes the observance and enforcement 



212 E. W. KELLOGG [J. S. M. p. E. 

of reasonable tolerances the more difficult. The large amount of 
literature on the subject bears testimony to the complexity of the 
problems. 

Experimental determinations of wave-shape distortion are of 
especial interest, and are undoubtedly a safer guide as to what may 
be expected than pure calculation. For such experimental results 
we may refer to the work of Messrs. Sandvik and Hall. 11 Of the 
numerous tests tabulated by these writers, a few of the best have 
been selected and are listed below. These may probably be safely 
taken as fairly representative of what may be expected with good 
processing control. The prints having a density of 0.60 made from 
negatives of density 0.75 correspond most nearly to what the authors 
refer to as normal conditions. 

TABLE I 

Tests of Variable-Density Recordings 

Light- R.M.S. 

Valve Output Harmonics 

Freq. Mod. Negative Print (Db. below (Per Cent of 

(Cps.) (Per Cent) Gamma Density Gamma Density 100%) Fundamental) 

100 90 0.5 0.55 2 0.40 7.1 8 



100 


90 


0.5 


0.55 


2 


0.60 


9.7 


11.5 


100 


90 


0.5 


0.75 


2 


0.40 


6.1 


9 


100 


90 


0.5 


0.75 


2 


0.60 


9.5 


13 


100 


50 


0.5 


0.55 


2 


0.40 


12.8 


5.6 


100 


50 


0.5 


0.55 


2 


0.60 


15.8 


5.6 


100 


50 


0.5 


0.75 


2 


0.40 


9.9 


6.25 


100 


50 


0.5 


0.75 


2 


0.60 


13.4 


6.0 



WAVE-SHAPE DISTORTION IN VARIABLE-WIDTH RECORDS 

Such distortion as takes place in variable-width records is of a 
different nature, and is due almost entirely to the spread of the 
exposure outside the boundaries of the exposed region and to the 
finite width of the recording slit. These two factors are similar 
in their effect. The ideal variable-width record would be one made 
with a slit of negligible width, and would have an infinitely sharp 
border between the exposed and unexposed areas. Actually, owing 
to the finite slit width, and to image spread, there is a gradation of 
exposure over a minute distance, and when recording such high 
frequencies that this distance becomes a considerable fraction of the 
wavelength, distortion becomes appreciable. With the exposures 
and developments used, the blackened areas are in general slightly 
oversized. Since both films are developed to fairly high contrast, 



Sept., 1935] COMPARISON OF SYSTEMS 213 

the edge of the image is, in general, sharper than that of the actual 
exposure. The spreading of the image in the negative is in a large 
measure compensated by similar spreading of the image in the print, 
and it has been found possible practically to neutralize distortion by 
proper choice of printing exposure. 

The effect of image spread has been analyzed by Cook, 12 the image 
spread being assumed for simplicity's sake to be due entirely to 
finite slit width. Cook's calculations show what may be taken as an 
upper limit or pessimistic estimate of the loss of fundamental, pro- 
duction of harmonics, and rectification or change in average trans- 
mission, as functions of the ratio of slit width to wavelength. The 
change in average transmission is, without doubt, the most serious 
of the three effects mentioned, because it results in difference tones 
which are well within the important audio range, whereas the loss 
of fundamental can within reason be compensated, and the har- 
monics generated are for the most part above the frequency range 
over which the reproducing system is sensitive. The justification 
for the last statement is that in the variable- width system the wave- 
shape distortion or generation of overtones is largely confined to 
the high-frequency waves for which the slit width is a considerable 
fraction of a wavelength. On the other hand, overtones produced 
in a variable-density system by non-linear film characteristics would 
be distributed throughout the audible range. The change in average 
transmission accompanies practically all non-linear distortion, and 
its magnitude closely follows that of the second harmonic produced. 
It is therefore a factor which may impair quality in either the vari- 
able-width or the variable-density system. Distortion which appears 
at high amplitudes, and at low as well as at high frequencies, is more 
likely to be injurious to musical reproduction where the peak powers 
are in the lower- and middle-frequency range than distortion which 
occurs only at high frequency where the waves to be recorded are 
of small amplitude. 

A further analysis of the effect of finite slit width in variable- 
width recording has been made by Foster, 13 who takes the grada- 
tions of exposure and the film characteristic into account. This, 
as would be expected, gives lower values for harmonics than Cook's 
assumptions. Foster points out that the same conditions which 
give freedom from distortion in the variable-density system, also 
eliminate the wave-shape distortion due to slit width in variable- 
width recordings, namely, a gamma product of unity and no ex- 



214 E. W. KELLOGG [J. S. M. p. E. 

posures below the straight parts of the H&D curves. This is not, 
however, the only way practically to eliminate this form of distor- 
tion. As has already been stated, suitable exposures of negative 
and print, without the requirement of unity gamma product, will 
accomplish the same purpose. High contrast and the lowest possible 
density in the clear areas are desirable in variable-width records, and 
neither of these is compatible with the conditions called for by the 
first method. Minimizing the distortion of high-frequency waves 
due to slit width and image spread is therefore accomplished by 
removing the causes, so far as possible, by the best of optical condi- 
tions and a very narrow recording slit, and then by proper printing. 
In the negative the dark areas are slightly oversize, but in the print- 
ing the exposure spreads slightly under the edge of the black parts 
of the negative, and by printing to a suitable density, the edge of 
the blackened area of the print can be brought back to where it would 
be had there been no spreading in either negative or print. 

TABLE II 

Variable-Width Records 

Freq. Negative 

(Cps.) Gamma Density 

100 2 1.3 

100 2 1.5 

100 2 2.1 

4000 2 1.25 

4000 2 1.46 

4000 2 2.1 

For an estimate of the magnitude of the distortion to be expected 
in a variable-width recording we may again refer to the valuable 
contributions of Sandvik, Hall, and Streiffert. 14 The values in 
Table II are taken from the Table II and the curves of Figs. 2 and 
3 of their paper on wave-form analysis of variable- width records. 14 
The output of fundamental frequency is expressed in decibels below 
100 per cent modulation of the incident light, and since the outputs 
of variable-density records have been given in the same terms, the 
figures are directly comparable. The modulation in the recording 
was 90 per cent. The harmonics are given in terms of the rms. 
amplitude of total harmonic content, in per cent of the amplitude of 
the fundamental. 

General recommendations for variable-width films call for a 
negative density of 1.4 to 1.5 and a print density of 1.3 to 1.4, and 



Print Density, 1.2 
Fundamental, Harmonics 
Db. below Per Cent 
100% of Fund. 


Print Density, 1.5 
Fundamental, 
Db. below 
100% Harmonics 


3.5 


3 


4.2 


4 


3.2 


5 


3.4 


4.6 


3.1 


7 


3.2 


6 


7.4 


3 


8.3 


5 


7.3 


4 


7.0 


2 


9.1 


7 


8.4 


5 



Sept., 1935] 



COMPARISON OF SYSTEMS 



215 



the figures given above show what may be expected in these ranges. 
The 2.1 density negatives are outside the range of ordinary use, but 
prints from these have been included in the list since some negatives 
may be denser than the next lighter ones listed. 

Comparing the harmonic distortion figures for the two systems 
(if we exclude the variable-width prints made from the 2.1 density 
negatives) we find that the variable-width records showed harmonic 
contents less than half of those of the 90 per cent modulated variable- 
density records, and materially less (relative to fundamental) than 
the variable-density records which were made with 50 per cent light - 



O .Z A A A 1.0 l.l I* *. t-ft 10 

PRINT DENSITY 

FIG. 1. Harmonic distortion in variable- width records at 100 
cycles per second as a function of print density for various nega- 
tive densities. Negative densities 0.8, 1.0, 1.25, 1.5, and 2.1 for 
curves A, B, C, D, and E, respectively. Negative gamma, 2.0. 

valve modulation. Since in the variable-density system distortion 
decreases at low amplitudes, while in the variable-width system 
amplitude is not a material factor, the superiority of the variable- 
width system on the score of wave-form distortion applies only to 
amplitudes in excess of 30 to 40 per cent. With amplitudes smaller 
than this neither system would produce sufficient distortion to be 
detected. 

The expectation of increasing distortion at high frequency in the 
variable- width system is not shown in the measurements cited here, 
but comparison of the curves of Figs. 1 and 2 which are reproduced 



216 



E. W. KELLOGG 



[J. S. M. p. E. 



through the courtesy of the authors of the paper cited, 14 shows 
that the range of negative and print densities which gives good results 
is less at 4000 cycles than at 100 cycles. Other factors than image 
spread must account for harmonics of the order of 3 or 4 per cent, 
since image spread could hardly have played an appreciable part at 
100 cycles. The printing operation was effective in neutralizing 
a large part of the image spread distortion, as evidenced by the fact 
that measurements by the same authors on the 4000-cycle negatives 
showed considerably larger harmonic ratios than those of the prints. 
At 100 cycles there was little difference between negatives and prints. 



18 



16 

I' 4 

jj 12 

g 
S' 
2 
i 

! 

H 4 

^ 
































\ 












i 












\ 


JC 








I 


1 













\ 


X 




, / 


















\ 


\ 
o 

\ 


V 


/ 




/ 


f 










N 


\r. 


^x 


7 


\ 


/ 


1 


/ 












\ 

/ 


4 


\ 


X, 


*. 




/ . 


















& 


/ 


\-, 


'K 




2. 

































t.Z 



1.6 I.& Z.O 



.Z A .< .O t.O 
PRINT 

FIG. 2. Harmonic distortion in variable- width records at 4000 
cycles per second as a function of print density for various nega- 
tive densities. Negative densities 0.85, 1.0, 1.3, 1.5, and 2.1 for 
curves A, B, C, D, and E, respectively. Negative gamma 2.0. 

In the experience in the Photophone laboratory at Camden, N. J., 
trouble from fogging-in of the waves does not begin until frequencies 
considerably above 4000 cycles are reached, and the measurements of 
harmonic distortion given above lead to a similar conclusion. 

COMPARISON OF OUTPUTS 

With respect to output, the 100-cycle, variable- width records are 
seen to give about 6 db. more output than the normal variable- 
density record, or from 2.5 to 3 db. more than the lighter print, which 
had a density of 0.4 for the unmodulated track. 



Sept., 1935] COMPARISON OF SYSTEMS 217 

A comparison of outputs at higher frequency can not be made, 
since the only measurements of variable- density records published 
were those at 100 cycles. Both systems are subject to loss of output 
with increasing frequency, and if the full potentialities of both sys- 
tems are attained, the high-frequency losses are at least of the same 
order of magnitude. 

GROUND-NOISE RATIOS 

In making any comparison of the two systems with respect to 
ground-noise we find it necessary to consider separately the ground- 
noise which arises from emulsion graininess in either film and from 
imperfections or specks on the negative, and that which is pro- 
duced by dirt and abrasions on the positive film. It will be con- 
venient to designate the former as type A and the latter as type B 
ground-noise. 

Little has been published in the way of actual measurements of 
type B ground-noise (dirt and scratches on the positive) and such 
noise, being more irregular than that due to graininess and depending 
as it does on accidental causes, is difficult to measure or estimate 
with sufficient accuracy to be significant. Theoretical considera- 
tions indicate that with respect to this kind of noise the variable- 
density records have an advantage. 

In order to illustrate the difference by a simple case, let us com- 
pare two films, both of which reduce the transmitted light to 50 
per cent, one by making half the track black, and the other by inter- 
posing a uniform gray across the whole track. If the light could be 
modulated 100 per cent in both films, the useful outputs would be 
the same and the same amplification would be required. Since 
the light-beam issuing from the transparent half of the variable- 
width track is twice as intense as that which has passed through 
the variable-density track (the same total light in a beam of half 
the cross-section) obstructions on that side of the track produce 
twice the light modulation, and therefore four times the energy in 
photo-cell output, that would be produced by identical obstruction 
which may be assumed to be found within the corresponding half 
of the variable-density track. But in the latter, the other half of 
the track makes an equal contribution to the noise. The noise 
generated on the two sides is, of course, in random relation, and under 
such conditions the total power is twice that produced by either 
half alone. This makes the total energy or power in type B ground- 



218 E. W. KELLOGG [J. S. M. p. E. 

noise twice as great for the variable-width track as for the variable- 
density track, or a difference of 3 db. If the average transmission 
is decreased by a uniform additional density, but the per cent modu- 
lation of such light as is transmitted is kept the same, the ratio of 
useful output to ground noise (of the type under discussion) is 
unaltered. The 3-db. advantage in type B ground-noise ratio of 
the variable-density record, is, therefore, not altered by the fact 
that average transmissions much below 50 per cent are employed. 
Neither is the type B ground-noise ratio for a variable-width record 
altered by a uniform fog or density over the transparent area, for, 
although this reduces the per cent modulation of the incident light, 
it does not reduce the per cent modulation of light reaching the 
photo-cell. The amplification would be turned up to compensate 
for the reduced illumination, and the useful sound and ground-noise 
would both be brought back to the original value. 

Whether the difference in type B ground-noise will be as great 
as the above-calculated 3 db., or something less, will depend on such 
questions as: 

(1) Can both systems provide equally high percentage modulation of the 
photo-cell light without excessive distortion? 

(2) Will the slightly rough or matte surface of a film having a density of 0.5 
or more tend to pick up or hold dirt more than the more glossy surface of clear 
film? 

(5) Is the gray film more or less easily scratched? 

(4) Will oil or wax tend to fill up and obliterate scratches more if they are in 
clear film than in gray film? 

Ground-noise reduction has been applied to both variable-density 
and variable-width records, the object being accomplished in both 
cases by reducing the average transmitted light during passages 
with low modulation. The variable-density records are made 
darker, while the width of clear track is reduced in the variable- 
width records. The same reasoning which we have already used 
shows that reducing the average transmission by darkening the 
track causes the ground-noise to fall off faster than an equal trans- 
mission reduction accomplished by narrowing the track. But in 
practice it has been found feasible to go further in the way of light 
reduction in the variable-width system, and a noise reduction of 
about 10 db. has been attained in both cases. There is a limit to 
how far it is desirable to go in any ground-noise reduction system of 
the variable-bias type. Fluctuating ground-noise is more con- 



Sept., 1935] 



COMPARISON OF SYSTEMS 



219 



spicuous than steady ground-noise, and the best degree of reduction 
is a compromise between the attainment of low ground-noise during 
soft passages and the avoidance of conspicuous fluctuations. This 
consideration is in part responsible for the practical limitation to 
10 db. 

With respect to type A ground-noise, or that due to emulsion 
graininess in both films, and to imperfections in the negative, the 
advantage is on the side of the variable- width system. The reason 
for this lies in the inherent difference between a system which depends 



-25 



-30 



-35 



d- 40 
I 

UJ-45 

<n 
6 

z 

-50 



-56 



O riNE. GRMN ElMUUblON 

MEDIUM SIT.E. GRfMN E.MUL.. 

X COARSE GRAIN E.MUV-SVON 




0.5 



1.0 



FIG. 3. 



SPECULAR DENSITY 

The relation between ground-noise and density. 



on blacks and whites, and one which depends on transmission through 
various shades of gray. The gray of the film is not uniform, especially 
when we are referring to microscopic areas such as fall within the 
reproducing light-beam. Despite the high degree of excellence of 
film emulsions, it could not be expected that the light which passes 
through a dense film would be as uniformly distributed as that which 
passes through a less dense film. This decreasing uniformity with 
increased density would be expected from considerations of proba- 
bility, but is borne out by experimental evidence. Were the per cent 



220 E. W. KELLOGG [J. S. M. P. E. 

variation in light from point to point the same for a dense film as 
for a lighter film, ground-noise due to film graininess or grain clump- 
ings would go down at the rate of 20 db. per unit increase in specular 
density. Curves published by Sandvik, Hall, and Grimwood 15 
and reproduced in Fig. 3 show that, starting with fresh unexposed 
film, the ground-noise increases with density up to a specular density 
of about 0.25; after which it falls, but not as rapidly as it would if 
the increased density merely cut down the light without affecting 
its uniformity. The maximum rate of decrease is about 15 db. 
per unit increase in specular density, and the difference between 
this and the 20 db. mentioned above is a measure of the increasing 
irregularity of light distribution at high densities. Since the pur- 
pose of a variable-density print is to reproduce all fluctuations in 
negative density, the graininess of the negative is necessarily printed 
through on the positive; and the positive, which is also a dark gray, 
adds its own graininess. Any scratches or specks which may be on 
the negative must likewise appear on the print. 

In the case of the black-and-white, variable-width records, the 
clear part of the positive is so far down on the toe of the film char- 
acteristic that its photographic sensitivity is practically zero, and 
it fails to register the irregularities of what little light reaches it 
through the dense part of the negative. Thus, negative graininess 
and specks* in the dark half of the negative do not show on the print 
or produce appreciable ground-noise; nor does the positive itself 
cause appreciable noise from graininess on the clear side, for where 
there are few grains there can not be graininess.** 

There remains to be considered the ground-noise which may be 
contributed by the black side of the print, whether resulting from 
its own graininess or from imperfections in the corresponding clear 
part of the negative. Here the total light transmitted is so feeble 
that even though it should be considerably modulated by imperfect 
distribution of the silver, the resulting noise is negligible. With a 
diffuse density of 1 .4, for example, which is about normal, the semi- 

* Scratches which leave actual holes in the negative emulsion would, of course, 
show as black spots in the print, but such injuries to films are rare. Nearly all 
scratches appear as dark lines. 

** Noise from graininess is not primarily the result of disturbances produced 
by individual grains, but of irregular distribution of the silver. Only when the 
number of grains becomes sufficient to produce material reduction of the trans- 
mitted light does graininess become appreciable. 



Sept., 1935] COMPARISON OF SYSTEMS 221 

specular density would be about 1.75 and the transmission to the 
photo-cell less than 2 per cent. 

Specks and scratches on the clear part of the negative would, of 
course, show as minute holes or clear spots in the otherwise black 
part of the print, but with the high exposure which this part of the 
print receives, these light spots tend to become fogged in because 
the specks and scratches are, for the most part, of very small 
dimensions. In this case, advantage is gained from the spread of 
exposure in the emulsion of the positive. Printing to a low density, 
as required for variable-density records, tends less to obliterate such 
marks. 

The essential difference between variable-width and variable- 
density records with respect to type A ground-noise may be better 
appreciated if we bear in mind that a variable-width track is made 
up of clear film, which has inherently low ground-noise, and of very- 
dark film which for a given amplifier setting also causes small ground- 
noise (under the conditions of its use in a variable-width track, in 
which the amplification required is dependent not on the transmission 
through the dark but through the clear side of the film). In the 
case of a variable-density track, on the other hand, clear film 
is not permissible on account of distortion, and whenever a high 
density is employed in the variable-density system, either the illumi- 
nation (in a printer) or the amplification (in a reproducer) must be 
increased correspondingly, and the full effect of the graininess of 
the dense film is registered. 

From the foregoing it appears that the relative advantage of one 
system or the other with respect to ground-noise will depend on 
the condition of the print and on which type of ground-noise is the 
more objectionable. If prints are in good condition, the variable- 
width records may be expected to have less ground-noise, especially 
the hiss type of noise characteristic of the film emulsion; whereas, 
as the print gets older the ground-noise due to scratches and dirt 
will probably increase more rapidly in films of the variable-width type. 

The hiss type of ground-noise has become a more serious problem 
since the introduction of reproducing equipment responsive to higher 
frequencies, and theaters which have the new equipment (pre- 
sumably, for the most part, first-run theaters) are benefited by the 
system which minimizes this type of ground-noise. The prints 
which they receive, being relatively fresh, should have comparatively 
little of the type B noise. 



222 E. W. KELLOGG [J. S. M. P. E. 

The advantage of the variable-width record in its relative freedom 
from ground-noise due to film grain commends it especially for the 
increasingly important application of master records to be used in 
re-recording. For this service the very minimum of ground-noise 
of all kinds would be sought, and ground-noise due to dirt and 
scratches is hardly a factor, because no print which had been in the 
least damaged would be employed. 

VISUAL MONITORING AND TRACK INSPECTION 

Factors which are of considerable practical importance are (1) the 
visual indication which the variable- width recording system provides 
of the amplitudes of the waves being recorded, and (2) the immediate 
judgment of a record which can be formed as soon as the negative 
can be examined. 

The relatively large amplitudes of movement of the light-spot as 
it plays across the slit enable the operator to judge how close it is 
coming to the overshooting limit, and thereby control his amplitudes 
with the maximum of nicety. Viewing screens are provided on 
which the proper limits of movement are marked, and as the screen 
is placed farther from the galvanometer than the slit, the movement 
is magnified. Alternative methods of observing amplitude, as, for 
example, by meter, do not reveal whether the modulating device 
is functioning, and may cause errors because of difference in the 
frequency characteristics of modulator and meter, failure of the 
meter to respond to peaks of short duration, or to the greater possi- 
bility of wrong adjustment when a separate indicating device is 
used. 

Inspection of a variable-width negative will reveal, almost without 
fail, whether the recording has been properly done, because the 
distinctive appearance of the waves enables a practiced eye to tell 
much about the characteristics of the entire recording system, and 
such factors as proper density, freedom from fogging, overshooting, 
and the maintenance of suitable levels are at once apparent. It 
would be interesting to know just how far an expert can go in inter- 
preting the appearance of a variable-density record, but certainly 
it is far more difficult than in the case of the variable-width record- 
ing, and rendered still more difficult by the fact that there are so 
many combinations of negative characteristics with various printing 
exposures and developments which give usable prints. Since play- 
ing a variable-density negative gives such high distortion that no 



Sept., 1935] COMPARISON OF SYSTEMS 223 

fair judgment could be formed by a listening test, final evidence that 
a recording has been satisfactorily made can not be had until a 
print has been made and played; whereas a variable-width negative 
may be played, and would in general be indistinguishable from a 
good print. 

Another factor which ought not to be a factor, but which un- 
fortunately is occasionally, is the effect of printer speed fluctuations. 
The output of a variable- width record is little affected by consider- 
able variations in recording light or printing exposure. There is, 
therefore, less modulation of the useful sound by recording or print- 
ing lamp fluctuations or by printer speed variations (due to gears 
or sprocket action) than would be the case with a variable-density 
recording. 

PUSH-PULL SOUND-TRACK 

The recently developed "push-pull" sound-track system which 
was described by G. L. Dimmick and H. Belar, 16 and first demon- 
strated before the Society at the Atlantic City Convention in April, 
1934, offers possibilities of a new standard in high-quality sound 
reproduction. Although its general introduction into theaters will 
necessarily require time, it is available now for master records. The 
outstanding advantage of the push-pull system is on the score of 
ground-noise, but it has another feature which deserves mention. 
We have discussed the wave-shape distortion and change in average 
transmission which occurs in recording high-frequency waves. The 
effect may be practically eliminated by proper printing, but in the 
push-pull system the change in average transmission and all the 
even harmonics are completely balanced out, thus removing from 
the variable- width system about the only photographic factor which 
is at all critical, and also carrying the already small distortion to a 
still lower point. 

But the really important feature of the push-pull track is that the 
track is black when there is no modulation, and the amount of clear 
film within the scanning beam at any time is proportional only to 
the instantaneous useful voltage being developed, instead of to the 
instantaneous value plus an average equal to the highest near-by 
peaks plus a substantial margin for safety, as is required in all present 
noise-reducing systems depending on bias. There is no margin 
needed to prevent overshooting or to enable the galvanometer or 
light-valve to open up in case of a sudden burst of sound, no sluggish 
closing down (as at present necessary to prevent a "pumping" 



224 E. W. KELLOGG [J. S. M. P. E. 

sound), and no rise and fall of the ground-noise imperfectly following 
the increase and decrease in modulation which is often noticeable 
in all systems depending on shifting bias. In the push-pull system 
the moment the output voltage drops the track is black again. 

These advantages of the new sound recording system were pointed 
out by Dimmick and Belar, but are restated here for the reason 
that the availability of this system is a factor to be weighed in 
comparing variable-density with variable- width. Although variable- 
density recordings may benefit in some measure from application of 
the push-pull principle, 17 the full advantages are realizable only 
if the print transmission can be brought practically to zero whenever 
the galvanometer current is zero (or the sound-pressure zero). For 
this, definite and reproducible linear relationship between galvanom- 
eter current and print transmission is essential, and it is in such 
linear relationship that the variable-width system has its great 
advantage. 

REFERENCES 

1 HARDY, A. C.: "The Rendering of Tone Values in the Photographic Re- 
cording of Sound," Trans. Soc. Mot. Pict. Eng., XI (1927), No. 31, p. 475. 

2 ZWORYKIN, V. K., HANNA, C. R., AND LYNN, L. B. : "The Kerr Cell Method 
of Recording Sound," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 35, p. 748. 

3 HARDY, A. C.: "Optics of Sound Recording Systems," Trans. Soc. Mot. Pict. 
Eng., XH (1928), No. 35, p. 760. 

4 KELLOGG, E. W.: U. S. Patent No. 1,740,406. 

5 DIMMICK, G. L.: "Galvanometers for Variable- Area Recording," /. Soc. 
Mot. Pict. Eng. t XV (Oct., 1930), No. 4, p. 428. 

6 SACHTLEBEN, L. T.: "Characteristics of the Photophone Light-Modulating 
System," /. Soc. Mot. Pict. Eng., XXV (Aug., 1935), No. 2, p. 175. 

7 JONES, L. A.: "Photographic Reproduction of Tone,". Opt. J Soc. Amer., 
V (May, 1921), No. 3, p. 232. 

8 TurrLE, C., AND MCFARLANE, J. W. : "The Measurement of Density in 
Variable-Density Sound-Films," /. Soc. Mot. Pict. Eng.,. XV (Sept., 1930), No. 
3, p. 345. 

MACKENZIE, D.: "Straight-Line and Toe Records with the Light- Valve," 
J. Soc. Mot. Pict. Eng., XVII (Aug., 1931), No. 2, p. 172. 

10 MEES, C. E. K.: "Some Photographic Aspects of Sound Recording," 
/. Soc. Mot. Pict. Eng., XXIV (April, 1935), No. 4, p. 285. 

11 SANDVIK, O., AND HALL, V. C.: "Wave-Form Analysis of Variable- Density 
Sound Recording," /. Soc. Mot. Pict. Eng., XIX (Oct., 1932), No. 4, p. 346. 

12 COOK, E. D.: "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV (June, 
1930), No. 6, p. 650. 

18 FOSTER, D. : "The Effect of Exposure and Development on the Quality of 
Variable-Width Sound Recording," /. Soc. Mot. Pict. Eng., XVII (Nov., 1931), 
No. 5, p. 749. 



Sept., 1935] COMPARISON OF SYSTEMS 225 

14 SANDVIK, O., HALL, V. C., AND STREIFFERT, J. C.: "Wave-Form Analysis 
of Variable- Width Sound Records," J. Soc. Mot. Pict. Eng., XXI (Oct., 1933), 
No. 4, p. 323. 

15 SANDVIK, O., HALL, V. C., AND GRIMWOOD, W. K.: "Further Investigations 
of Ground-Noise in Photographic Sound Records," J. Soc. Mot. Pict. Eng., 
XXII (Feb., 1934), No. 2, p. 83. 

16 DIMMICK, G. L., AND BELAR, H.: "An Improved System for Noiseless Re- 
cording," J. Soc. Mot. Pict. Eng., XXIII (July, 1934), No. 1, p. 48. 

17 SHEARER, D.: "A Variable- Density Recording Method to Produce In- 
creased Undistorted Volume Range." Presented at the Spring, 1935, Meeting 
at Hollywood, Calif. 

DISCUSSION 

DR. J. G. FRAYNE : It is true that the lowest density of the straight-line portion 
of the characteristic is about 0.6, which would indicate that that would be the 
lowest density that would be practicable if we want to adhere to the straight-line 
portion of the positive film. We have found that a density of 0.7 in the unbiased 
portion, corresponding to about 20 per cent visual transmission, represents a 
fair value of print density from the standpoint of low harmonic content and 
faithfulness of wave-shape. The practice in Hollywood varies considerably from 
studio to studio. The darkest prints made are about 0.8 in visual density, and 
vary all the way to 0.5, with the average probably around 0.65, or 22 per cent. 

Now, if we take a variable-density film of an average transmission of 22 per 
cent, we shall find the output from that film to be approximately 7 or 8 db. below 
the average of variable-width films. However, those of us who belong to the 
variable-density school do not regard that as any particular disadvantage. It 
simply means a change in the fader setting, and since films are mixed at such 
vastly different levels, the fader correction has to be made anyhow, in order to 
obtain uniform reproduction of sound in the theater. However, certain studio 
executives whose ears are not the very best insist upon a certain sound volume 
from some fixed fader setting. Thus, certain sound directors, against their 
wishes, are forced to make prints lighter just to satisfy their immediate executives. 
However, the studios that make lighter prints make them not on the basis of 
data from sensitometric studies, but on the basis that they sound better than 
the darker prints. I have been sceptical of this for a long time, but recently 
have begun to change my mind somewhat. A series of exposures made very 
close together and developed in machines in which there are pronounced di- 
rectional affects, may give widely different results from a scheme in which the 
densities are scattered over a long band of film, as in a sound-track. That is 
borne out by the test known as the delta db., discussed by Mr. Albin before this 
Convention the other day. Very often the Eastman 2B data will tell us that 
we should make a variable-density print at 0.7 for the best results, and yet the 
dynamic test will indicate that the density should be 0.5 or 0.6, and in every case 
this conclusion appears to be borne out by listening tests. So, until we know 
more about how to measure densities, or until we get a sensitometer that will really 
tell us the physical facts, our ears will have to give us the final verdict. Based 
upon the ear, the optimal print may vary from 15 to 30 per cent. 



226 E. W. KELLOGG 

Despite the fact that the harmonics may appear to increase rapidly with change 
of density, it is customary practice in several studios to make use of a fairly wide 
portion of print transmission to secure volume control. It is generally recognized 
that a density range from 0.8 up to possibly 0.5 can be used without any serious 
deterioration of sound. That gives a 6-db. volume range. I do not believe such 
control by print density in the variable-width system is practicable without 
introducing other difficulties. The ability to control print volume in the variable- 
density system by printer light control is an outstanding advantage. 

MR. KELLOGG: Dr. Frayne has raised the question of the possibility of con- 
trolling output levels by changing the print exposure. If certain passages have 
been recorded at higher levels than desired there are two methods by which they 
may be reduced : (1 ) re-recording and (2} dark printing. In the case of re-record- 
ing, the light-valve would be given small modulation and bias. This results in a 
light negative and a dark print. The method of increasing the printer light, 
using the original negative, likewise gives a dark print. I have pointed out in the 
paper that if a given print density is obtained by high printing light through a 
dense negative, the resulting graininess in the print is greater than would be the 
case with less printer light and a thinner negative. It should be possible to ob- 
tain about the same percentage modulation of the light reaching the photo-cell 
whichever way the print is made. The re-recording system, therefore, offers the 
possibility of the same volume of useful sound with less hiss due to film grain. 
The ground-noise due to dirt and scratches upon the print is the same for the two 
types of printing, since both are assumed to have equal average density. 

Since re-recording offers a preferable means of obtaining the reduced-level print, 
it would appear that the application of the heavy -print method should be limited 
to cases in which the number of prints to be made is not sufficient to justify a 
re-recording operation. If re-recording is to be employed we are brought back 
to the question whether variable-density or variable-width recording will give 
the best results (Jf) for the original and (2) for the reduced-level recording. 

If the number of prints to be made is small, and re-recording therefore not justi- 
fied, both the variable-density and the variable-width systems permit reducing the 
level of certain passages by a printer operation. In the variable-density system 
the printing light is increased and the result is a dark print. In the variable- 
width system the objective can be attained by flashing the print, or, in other 
words, by giving the entire sound-track an exposure in addition to that which 
prints the sound. This will result in a uniform gray in the transparent areas, while 
the density in the dark areas should be kept about the same as for a standard print. 
Tests have been made which show that a reduction of the order of 6 to 8 db. can 
be made by the flashing operation, without appreciable impairment of the relative 
high-frequency output or other distortion. The additional exposure may be at- 
tained by running the film through the printer a second time if only a few prints 
are wanted. For a larger number an auxiliary lamp can be readily arranged. 

Assuming for comparison that the standard prints would give equal ground- 
noise, the reduced-level print would have less hiss in the variable-width system. 
The reason for this is that in the case of a variable-density print, the darkening of 
the print is attained by a strong light through a comparatively dense negative, the 
graininess of which is added to that of the positive; whereas in the variable-width 
system only the inherent graininess of the positive enters. 



RELATION BETWEEN ILLUMINATION AND SCREEN SIZE 
FOR NON-THEATRICAL PROJECTION* 

D. F. LYMAN** 

Summary. Methods of measuring the illumination output of 8 -mm. andl6-mm. 
projectors are discussed, with particular reference to some of the many variables that 
affect the results. This output value, expressed in screen lumens, permits determining 
the screen size for any desired foot-candle level. But because screen brightness de- 
pends also upon the reflection characteristics of the screen material, there is presented 
a classification of various types of screen material. For each class of screen, maxi- 
mum and minimum illumination values are suggested, with the object of keeping the 
screen brightness within the limits necessary for good picture quality. Charts illus- 
trate the relation between these factors and between the screen size and projection 
distance. 

During the development of 8- and 16-mm. projectors, the screen 
illumination has been increased steadily by the adoption of more and 
more efficient optical systems and lamps of higher wattage. This 
increase has introduced another projection problem, a tendency to- 
ward excesssive screen brightness, which may be nearly as objection- 
able as under-illumination. It is the purpose of this paper to suggest 
flexible limits for screen brightness, taking into consideration the 
light flux from the projector and the size and reflection character- 
istics of the screen. 

LIGHT FLUX FROM THE PROJECTOR 

The luminous flux from a projector is expressed in terms of screen 
lumens. One method of determining its value is as follows : The 
projector is run at about normal speed without film, and the light 
beam is projected to a wall or screen upon which can be measured 
the magnified image of the gate aperture. For this purpose the lens 
is focused so that the image is sharply defined, and the aperture is 
framed properly. Since the illumination varies slightly with the 
position of the lens, the image is made about average screen size, 
at least three or four feet wide. Then, by means of an illuminometer, 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Eastman Kodak Co., Rochester, N. Y. 

227 



228 D. F. LYMAN [J. S. M. P. E. 

readings in foot-candles are taken at nine or more points in the field. 
If the points are well distributed over the field, including the corners, 
the readings at once show the degree of uniformity of the illumina- 
tion. Their average multiplied by the area of the field in square 
feet gives the flux, F, in screen lumens, 

F = 0.75 W*E 

where E is the average illumination in foot-candles and W is the 
width of the screen in feet. The 0.75 factor is the ratio of aperture 
height to width. 

Corrections must be applied in order to attain consistent results, 
one of which is for the variation of area of the lamp filaments. 
Source areas of a number of lamps of the same type are measured, 
and the dimensions are checked against the specified size. Then 
illumination tests made with the same lamps reveal the variation 
that may be expected with various filament spreads. 

Another reason for including several lamps is the slight variation 
in the location of filaments with respect to the optical axis. Pre- 
focus-based lamps have eliminated most of the uncertainty about 
centering, but there are still slight mechanical tolerances that require 
particular attention. Instructions furnished with projectors describe 
the method of adjusting lamps and reflectors, and also stress the 
importance of keeping all glass surfaces clean. In illumination tests 
these rules must be followed. 

A third reason for averaging results from a number of lamps is the 
slight deviation from normal filament temperature. Lamp manu- 
facturers attempt to keep this factor constant so that the design 
life will be attained. There is, however, enough variation between 
lamps to necessitate averages. New lamps sometimes show a slight 
increase of output for the first half or three-quarters of an hour; 
after which the illumination decreases gradually to about 75 per cent 
of its initial value before the filament fails. This decrease is caused 
by changes in the filament and by blackening of the bulb. Curves 
of illumination plotted against time are valuable because they show 
initial illumination, the rate of decrease of illumination, and the lamp 
life. Furthermore, if a sufficient number of such illumination main- 
tenance curves are prepared, it is possible to estimate the probable 
performance of a lamp in question if only its initial illumination is 
read. No measurement of projector performance is complete without 
correcting for the characteristics of the lamp. Lamps that have been 



Sept., 1935] ILLUMINATION AND SCREEN SIZE 229 

selected for correct source area and rated in volts or amperes for the 
proper filament temperature can be procured from the lamp manufac- 
turers. Ordinarily, however, results from the run-of-the-mill product 
are of chief interest. 

In addition, dependable results require the use of an accurate 
voltmeter or ammeter, depending upon whether the lamp is rated 
in volts or amperes. Because the present high-powered lamps draw 
a fairly heavy load and often cause a drop in the line voltage, re- 
sistance adjustments are made and meter readings are taken only 
when the full projector load is connected across the line. Photom- 
eters of all types require frequent checking, especially if the fila- 
ment temperature of the projection lamp differs from the temperature 
for which the illuminometer was calibrated. 

When the output of a large number of projectors must be ascer- 
tained, an integrator-box with a ground-glass or opal-glass screen is 
placed in front of the projection lens. At the other end of the box is 
inserted the light-sensitive cell. Or, if a photometer of the visual 
type is employed, a diffusing glass is placed in each end of the box 
and readings are taken from the glass farthest from the projection 
lens. With either type of photometer, readings from the box are 
compared with values obtained by the conventional point-by-point 
method at the screen, and a conversion factor is established. Inser- 
tion of neutral gray filters of the proper density permits reading from 
the box directly in screen lumens, although the illuminometer may 
be calibrated in foot-candles. One objection to the integrator-box 
method is the impossibility of measuring uniformity over the screen 
area. On the other hand, it is especially useful when several lamps 
or lenses of different types are under test. 

Some laboratories use more complicated equipment, such as 
hemispheres into which the beam is projected. Such equipment 
enables skilled operators to read illumination with great precision, 
but the ordinary methods just described are accurate enough for 
this purpose if proper corrections are made. 

Tests are usually made with the shutter operating, and this should 
be stated when the results are given. Except for still-picture projec- 
tion, the light that passes the rotating shutter is of major interest. 
Occasionally, however, the efficiency of only the optical system is 
desired, in which case both the heat screen and the main shutter are 
blocked open and kept stationary. 



230 



D. F. LYMAN 



[J. S. M. p. E. 



METHODS OF EVALUATING PROJECTOR PERFORMANCE 

Expressing projector performance in screen lumens is the real 
criterion, as opposed to the common practice of judging according to 
lamp wattage. If all lamps, optical systems, and shutters had the 
same efficiency, the latter method would be practicable. But a 
comparison of the ratio of screen lumens to lamp wattage for a large 
number of projectors of different types reveals a surprising discrep- 
ancy in efficiency. Twenty-eight typical projectors with regular 
lens equipment average 0.21 screen lumen per watt, but the maxi- 
mum is 0.43 and the minimum is 0.07. Although this difference of 
6 to 1 is an extreme case, factors of 3 to 1 and 2 to 1 are quite com- 
mon. Low- voltage lamps with filaments offset toward the condenser 
are generally high in efficiency, because both the size and position 
of the filaments permit the condensers to collect larger cones of light. 
From an efficiency standpoint, projectors with biplane filament lamps 
are not necessarily superior to those with monoplane filaments. But 
the type of lamp is only one factor in the final efficiency result; the 
others are the design and spacing of the condenser and the reflector 
and the transmission of the shutter and projection lens. Much of the 
variation in efficiency can be charged to these collecting and trans- 
mitting parts of the projectors. 

TABLE I 
Range of Efficiency with Shutter Transmission of 60 Per Cent 



Objective 


Screen Lumens 
per Watt 


Screen Lumens; Shutter Running 


400-W. Biplane 


500-W. Biplane 


750-W. Biplane 


//i.e 

//2.0 
//2.5 


0.25 to 0.30 
0.20 to 0.25 
0.15 to 0.20 


100-120 
80-100 
60-80 


125-150 
100-125 
75-100 


187-225 
150-187 
112-150 



As biplane-filament lamps have been adopted extensively, there 
is shown in Table I the range of efficiency that may be expected 
from a well-designed 16-mm. projector having a shutter transmission 
of at least 60 per cent. Table I is given merely as an illustration of 
favorable design conditions. The figures, therefore, should not be 
applied indiscriminately to any projector with the equipment tabu- 
lated. 

Rather wide limits of efficiency and luminous flux are given be- 
cause not all objectives with the same // value have the same trans- 



Sept., 1935] ILLUMINATION AND SCREEN SIZE 231 

mission. Moreover, there is a slight decrease of efficiency with an 
increase of wattage of the lamp and the size of the source. 

Eight-mm. projectors of different types have delivered 0.02 to 
0.15 screen lumen per watt. Here again, the maximum is attained 
with offset-filament construction. With centered monoplane lamps 
and favorable design, 0.06 to 0.08 screen lumen per watt represents 
average efficiency, whereas biplane filaments give 0.07 to 0.09. 

Classifying projectors according to luminous flux output has an 
additional advantage. It is quite easy to determine the screen size 
for an assumed illumination, or the illumination for a given screen 
size by rearranging the formula: 



to W = -V//TS.5 or E = 



0.75W 2 
CLASSIFICATION OF SCREENS 

Before proceeding with recommendations for screen brightness, 
it will be necessary to consider the reflection characteristics of the 
several kinds of screen materials. 

Class 1 (Fig. 1) is an arbitrary designation for a matte or diffuse 
screen, a surface that appears equally bright through a viewing angle 
of at least 30 to 40 degrees from the optical axis. This surface should 
be as white as possible, especially for color pictures. Its reflection 
factor should be at least 70 per cent. A white unglazed paper 
similar to typewriter paper is a good example of this class. 

Class 2 is a semi-matte screen, which, through an angle of plus 
and minus 30 degrees from the optical axis, has an average reflecting 
power about twice as great as that of a Class 1 screen. Slightly 
glossy white surfaces, some beaded screens, and rough surfaces with 
an aluminum coating fall into this class. 

Class 3 is a semi-specular screen, which, as the number indicates, 
has an average reflecting power about three times as great as that 
of a Class 1 screen, but only through a 60-degree angle. Many 
aluminum coated surfaces of a fairly fine texture are included in this 
group. 

Class 4 is about as specular as can be used for projection. It 
has a high reflection factor on the axis or at an angle of reflection equal 
to the angle of incidence, but the brightness decreases rapidly until 
at 30 degrees it is less bright than a Class 1 screen, assuming that 
both classes are subjected to the same illumination. This kind of 



232 



D. F. LYMAN 



[j. a M. p. E. 



screen is used when maximum brightness is demanded, as for Koda- 
color film and 8-mm. projectors in the lower price-class. Smooth 
metal, cardboard, or paper surfaces that have been coated with a 
specular aluminum finish are included in this class. 

Fig. 1 delineates characteristics of the four classes. The curves 



800 




5 

cr 



30 



20 10 10 20 

DEGREES FROM MAXIMUM 



30 



FIG. 1. Characteristics of the four classes of screens. 



represent averages of several materials in each group. A block of 
magnesium carbonate, represented by 100 upon the vertical scale 
and taken as the reference standard, would be the most efficient 
Class 1 surface, but since its use as a screen material is not practi- 
cable, the Class 1 curve is shown somewhat below the 100 line. 



Sept., 1935] ILLUMINATION AND SCREEN SIZE 233 

LIMITS OF SCREEN BRIGHTNESS 

Screen brightness could be expressed in absolute terms of milli- 
lamberts or candles per square foot, but these units become confusing 
when changing from diffuse to specular screens. Instead, it is simpler 
to state the limits in terms of illumination, shutter running but 
without film, for each class of surface. For example, with a Class 1 
screen the suggested upper limit for illumination is 24 foot-candles. 
As Kodacolor and Kodachrome present different problems, they are 
treated separately later. With a Class 1 screen an illumination of 
more than 24 foot-candles is undesirable for the average black-and- 
white film, whether it be reversal, duplicate, or reduction print. In 
the first place, most persons agree that excessive brightness detracts 
from the pleasing quality of the denser portions of the picture. 
Shadows become light gray and graininess in medium densities be- 
comes more noticeable. In some cases observers have noted an 
apparent spreading or enlarging of the highlights. Furthermore, 
nicker is objectionable, even though most 'projectors have 48 light 
interruptions per second at normal projection speed. If there is a 
slight travel-ghost, it is very apparent at high illumination levels, 
while with less brightness it may not be visible. 

To keep the maximum brightness constant for the four classes of 
screen material, upper limits of 12, 8, and 6 foot-candles are sug- 
gested for Classes 2, 3, and 4 screens, respectively. 

With a Class 1 screen there is a wide range of illumination from 24 
down to about 4 foot-candles, which allows good projection quality. 
Changes in the eye, the increase in the size of the pupil and the 
greater sensitivity at lower levels of brightness tend to maintain 
pleasing quality. Below 4 foot-candles, however, quality again be- 
gins to suffer. The usual reaction is that medium densities appear 
quite black and lose detail, and highlights are weak and unreal. 
Yet, under some conditions results are passable with only 1 foot- 
candle. Scenes including a preponderance of highlights, overex- 
posed scenes, light prints, and cartoons may appear bright enough to 
justify this low value. 

Conversely, films that are dark, whether from the nature of the 
subject, underexposure, or heavy printing, may warrant raising the 
upper limit of 24 foot-candles. In neither case should the limits be 
considered inflexible. 

Another factor that affects the minimum figure is the level of 
room illumination. For most 16- and 8-mm. projection the room is 



234 



D. F. LYMAN 



[J. S. M. P. E. 



dark except for stray light from the projector and light reflected from 
the screen. If there is too much added light, for example, when 
projecting classroom films in the daytime, the 4 foot-candle minimum 
may be too low. If it is necessary to project a dim picture, results 



400 




15 



2 3 4567 8910 
WIDTH OF SCREEN - FEET 



6 20 25 



10 



FIG. 2. Chart for determining screen size for 16-mm. projection, with sug- 
gested limits of screen illumination. 

are better if the room is kept dark for at least 10 minutes before 
projection to allow the eyes to become dark-adapted. 

Minimum values for Class 2, 3, and 4 screens are placed at 3 foot- 
candles. This might be lower for narrow rooms that would require 



Sept., 1935] ILLUMINATION AND SCREEN SIZE 235 

grouping the spectators in an angle of plus and minus 15 degrees 
from the optical axis. But when large screens are used, many of the 
spectators are usually relatively close to the screen and view it from 
a considerable angle. At 30 degrees all four classes of screens are 
about equal in reflection, while at 20 degrees even Classes 3 and 4 
are only about twice as bright as a Class 1 screen. Suggested limits, 
then, are grouped in Table II. 

TABLE II 

Suggested Limits of Illumination 
(Foot-Candles) 

Shutter Running, No Film in Gate 
Class of Screen Minimum Maximum 

1 4 24 

2 3 12 

3 38 

4 36 



Kodacolor Film requires special consideration because the absorp- 
tion of the filter and the limiting/ value of the optical system reduce 
allowable screen size. When Kodacolor was introduced, proper 
color saturation dictated the use of a 16V2 by 22-inch screen with 
the illumination afforded by the most efficient projectors available 
at that time. Since then, the gain in illumination and the substitu- 
tion of filters having higher transmission factors have permitted 
increasing the screen size successively to 22 by 30 inches, 30 by 40 
inches, and even larger for some of the efficient projectors. Thus, 
one of the chief obstacles to successfully viewing Kodacolor was 
removed. In general, this color process requires maximum screen 
brightness, which means specular screens, high-powered lamps, and 
fast lenses. 

Projection of Kodachrome, the new full-color film, is not handi- 
capped by lenticular embossings or by projection filters. Here the 
color is in the film itself. Instructions state that Kodachrome can 
be projected to the same size as black-and-white. Actually, how- 
ever, the minimum illumination level can not be stretched so much for 
Kodachrome as for black-and-white, because the colors lose their 
brilliance. For this film, then, the lower limits suggested in Table II 
are as low as should be employed. With a normal screen size of 
about 39 by 52 inches and a projector output of 50 to 200 screen 
lumens, there need be no special treatment for Kodachrome. In 



236 



D. F. LYMAN 



[J. S. M. p. E. 



* *> 

N 

& 



tf c* ' V" ^ <l 

* ///// 



FOOT CANDLES 



fc ^ V > <*/ 







15' 



'I 15 2 3 45678910 

WIDTH OF SCREEN -FEET 

FIG. 3. Chart for determining screen size for 8-mm. projection, 
with suggested limits of screen illumination. 



Sept., 1935] 



ILLUMINATION AND SCREEN SIZE 



237 



fact, it can be spliced into black-and-white films and projected under 
the same conditions with good results. 

Except for Kodacolor and large black-and-white screens, then, 
illumination requirements are not high. Screen size is often fixed 
by the length of the room and the focal lengths of the lenses avail- 
able. If this is true, there are two simple ways to control screen 

TABLE III 

Relation between Screen Lumens and Screen Size 
(For Both 16- Mm. and 8-Mm. Projectors) 



Screen 
Lumens; 
Shutter 
Running 


Width of Screen in Feet 


Class 1 
Min. Max. 
4 Ft.-c. 24 Ft.-c. 


Class 2 
Min. Max. 
3 Ft.-c. 12 Ft.-c. 


Class 
Min. 
3 Ft.-c. 


3 
Max. 
8 Ft.-c. 


Class 4 
Min. Max. 
3 Ft.-c. 6 Ft.-c. 


4 


1.2 


0.5 


1.3 


0.7 


1.3 


0.8 


1.3 


0.9 


8 


1.6 


0.7 


1.9 


0.9 


1.9 


1.1 


1.9 


1.3 


12 


2.0 


0.8 


2.3 


1.1 


2.3 


1.4 


2.3 


1.6 


16 


2.3 


0.9 


2.7 


1.3 


2.7 


1.6 


2.7 


1.9 


20 


2.6 


1.0 


3.0 


1.5 


3.0 


1.8 


3.0 


2.1 


24 


2.8 


1.1 


3.3 


1.6 


3.3 


2.0 


3.3 


2.3 


28 


3.0 


1.2 


3.5 


1.8 


3.5 


2.2 


3.5 


2.5 


32 


3.3 


1.3 


3.8 


1.9 


3.8 


2.3 


3.8 


2.7 


36 


3.5 


1.4 


4.0 


2.0 


4.0 


2.5 


4.0 


2.8 


40 


3.7 


1.5 


4.2 


2.1 


4.2 


2.6 


4.2 


3.0 


50 


4.1 


1.7 


4.7 


2.4 


4.7 


2.9 


4.7 


3.3 


75 


5.0 


2.0 


5.8 


2.9 


5.8 


3.5 


5.8 


4.1 


100 


5.8 


2.4 


6.7 


3.3 


6.7 


4.1 


6.7 


4.7 


125 


6.5 


' 2.6 


7.5 


3.7 


7.5 


4.6 


7.5 


5.3 


150 


7.1 


2.9 


8.2 


4.1 


8.2 


5.0 


8.2 


5.8 


175 


7.6 


3.1 


8.8 


4.4 


8.8 


5.4 


8.8 


6.2 


200 


8.2 


3.3 


9.4 


4.7 


9.4 


5.8 


9.4 


6.7 


225 


8.7 


3.5 


10.0 


5.0 


10.0 


6.1 


10.0 


7.1 


250 


9.1 


3.7 


10.5 


5.3 


10.5 


6.5 


10.5 


7.5 


275 


9.6 


3.9 


11.1 


5.5 


11.1 


6.8 


11.1 


7.8 


300 


10.0 


4.1 


11.5 


5.8 


11.5 


7.1 


11.5 


8.2 


325 


10.4 


4.2 


12.0 


6.0 


12.0 


7.4 


12.0 


8.5 


350 


10.8 


4.5 


12.5 


6.2 


12.5 


7.6 


12.5 


8.8 


375 


11.2 


4.6 


12.9 


6.5 


12.9 


7.9 


12.9 


9.1 


400 


11.6 


4.7 


13.3 


6.7 


13.3 


8.2 


13.3 


9.4 



brightness: (1) by changing to a higher- or lower- wattage lamp, or 
(2) by using a more diffuse or more specular screen. Obviously, it is 
good economy to use a lamp of low wattage, principally because the 
initial cost is lower and secondarily because operating expense and 
heat are decreased. If a high-wattage lamp is desired for Kodacolor 
or for occasional showings with a large screen, excessive screen bright- 



238 D. F. LYMAN 

ness with a smaller screen will be avoided by utilizing a diffuse sur- 
face. To the advantage of uniform brightness through a wide angle 
there is added the freedom from texture often visible with specular 
screens. In some cases it is best to use both means of reducing the 
screen brightness if the size must remain constant. 

Determination of screen size from Fig. 2, which is for 16-mm. 
film, involves a knowledge of the number of screen lumens. From 
the corresponding point upon the left vertical scale, follow a hori- 
zontal line to the desired foot-candle line, which can be selected after 
consideration of the class of screen and the illumination limits given 
for each class. From the intersection of the screen lumen and the 
foot-candle lines follow a vertical line to the bottom horizontal scale 
and read the width of the screen in feet. The projection distance 
for any screen width can be found by using the second set of lines, 
labelled with the focal lengths of the projection lenses. In this case 
the right vertical scale shows the throw in feet. Fig. 3 is similar 
but plotted for 8-mm. projectors. 

Table III reveals the same relationship between screen lumens 
and screen size, but here the range in screen width for the suggested 
limits of illumination is given. This table applies to both 16- and 
8-mm. projection. 

Although it is very difficult to make hard and fast rules for limit- 
ing screen brightness, an attempt has been made here to outline some 
of the factors that must be considered if the sensation as produced 
by projection is to be commensurate with the fine quality that is 
inherent in correctly exposed films. 



SENSITOMETRIC STUDIES OF PROCESSING CONDITIONS 
FOR MOTION PICTURE FILMS* 

H. MEYER** 

Summary. Sensitometric and pictorial tests made and processed in eight major 
Hollywood motion picture laboratories are discussed with special reference to the 
characteristics of the printed over-all curves. Attention is called to the importance of 
keeping the difference between contrast and gamma distinct. Several theoretical and 
practical conclusions resulting from the study of these tests are formulated. 

INTRODUCTION 

The following is a report of a series of tests conducted in eight 
Hollywood motion picture laboratories for the purpose of studying 
the various conditions under which 35-mm. film is processed. From 
the data obtained some conclusions were formulated which will be 
presented later. This is done not with the intention of criticizing 
existing laboratory practice but rather with the earnest desire to con- 
tribute in a modest degree to the understanding of photographic laws 
and requirements so essential to continuously maintaining standard 
quality. 

TEST METHODS 

The tests were prepared with the thought in mind of checking by 
sensitometric control every phase of processing the pictorial record 
obtained in the production of motion pictures. For this purpose a 
roll of supersensitive panchromatic gray-back negative film was ex- 
posed in a Bell & Howell camera, on a title board, photographing a 
vase placed in front of a black-and-white split background. Lighting 
and exposure were carefully determined by preliminary tests, taking 
into consideration speed and contrast of this particular negative film 
type so as to assure a fully exposed negative with normal contrast 
under average developing conditions. The object was illuminated by 
Mazda light. The split background consisted of black velvet and 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Agfa Ansco Corp., Pacific Coast Technical Division, Hollywood, Calif. 

239 



240 H. MEYER [J. s. M. P. E. 

white glossy cardboard, both areas large enough to permit reading 
their respective densities in the developed image by a Capstaff densi- 
tometer. 

The exposed roll of negative was broken into eight parts which 
were in turn submitted to each laboratory for development under 
normal production conditions. Each roll was accompanied by a 
sensitometric strip, exposed in an Eastman time-scale sensitometer, 
on film from the same roll. 

When the processed negatives were returned the sensitometric 
strips were read and plotted in the usual manner and, in addition, the 
two density values of the split background in all pictorial negatives 
were determined. These figures were then marked upon the corre- 
sponding sensitometric graphs, thus indicating quite accurately the 
position of the lowest shadow detail and the strongest highlight in the 
object. A loop was then made from each pictorial negative and its 
corresponding sensitometric strip, and returned to each laboratory 
with the request that it be printed upon black-and-white positive 
film in one of their regular Bell & Howell production printers, on all 
odd numbered lights from 1 to 21. These prints then went through 
the normal positive developing procedure at each laboratory, accom- 
panied by a sensitometric strip exposed on the same type of positive 
film in an Eastman time-scale sensitometer, in order to obtain a 
record of the positive solution characteristics. All sensitometric 
strips representing over-all prints from the original negative strip 
over the full range of the model D Bell & Howell printer were then 
read and plotted in conjunction with the negative graphs. Again the 
density values of the split background were determined in each print 
and marked upon the corresponding over-all curves. In addition, the 
sensitometric positive solution strips were read and plotted separately. 

TEST RESULTS AND CONCLUSIONS 

Listed in Table I are the gamma values obtained from each labora- 
tory, including negative gamma, positive solution gamma, printed 
over-all gamma, and over-all gamma as obtained by multiplying the 
negative gamma by the positive solution gamma. 

The difference between the printed over-all gamma and the over-all 
gamma obtained by multiplying the negative and positive solution 
gammas is also listed and expressed in percentage. From this table 
one will notice that the average negative gamma is close to 0.67 and 
that the largest deviation from this value is +0.04 and 0.05. The 



Sept., 1935] PROCESSING CONDITIONS 241 

differences between the positive solution gammas as maintained in 
these laboratories is considerably larger, the average gamma being 
2.37, with deviations as high as +0.31 and 0.37. The range of 
values for the printed over-all gammas ranges from 1.18 to 1.62, with 
an average value of 1.43, and deviations of +0.19 and 0.25. The 
printer factor was found in every instance to be close to 1 1 per cent. 

TABLE I 

Positive Printed Over-All Gamma Printer 

Negative Solution Over- All (Gamma Negative X Factor 

Laboratory Gamma Gamma Gamma Gamma positive) (per cent) 

1 0.71 2.00 1.18 1.42 12 

2 0.65 2.15 1.25 1.40 11 

3 0.62 2.34 1.34 1.45 11 

4 0.70 2.23 1.46 1.61 11 

5 0.66 2.40 1.48 1.58 11 

6 0.68 2.48 1.53 1.68 11 

7 0.62 2.68 1.54 1.66 11 

8 0.67 2.66 1.62 1.78 11 

All pictorial prints were submitted to the judgment of several 
timers, who selected the most satisfactory print from each laboratory. 
Listed in Table II is the printing light corresponding to each selected 
print. Also listed is the density range for each selected print which 
was determined by the readings of the two density values of the split 
background, and which were marked upon the corresponding over-all 
curves by A and B. In the fourth column will be found the globe 
characteristics of each printer lamp. 

TABLE II 

Printing Density Range 

Laboratory Light (A-B) Printer Lamp 

1 19 0.24-1.62 100-watt, outside frosted 

2 13 0.27-1.66 100-watt, outside frosted 

3 13 0.27-1.73 75-watt, inside frosted 

4 17 0.14-1.82 100-watt, clear 

5 13 0.23-1.69 60-watt, inside frosted 

6 19 0.24-1.64 60-watt, inside frosted 

7 11 . 23-1 . 78 75-watt, inside frosted 

8 13 0.22-1.84 75-watt, inside frosted 

Readings of the density range for laboratory No. 4 are not quite 
satisfactory for the reason that, due to a misunderstanding, the nega- 
tive loop was printed only up to light 17 while a higher light would 
have given a better print. The variations of contrast between the 
selected prints from each laboratory were quite pronounced. 



242 H. MEYER [J. S. M. P. E. 

Figs. 1 to 7 show the sensitometric graphs obtained from each 
laboratory. 

The small letters a and b mark the two densities of the negative 
background, while the capital letters A and B mark the corresponding 
densities of the printed background. Lines were also drawn parallel 
to the density axis through the toe and shoulder breaking-points of 
the negative curves and parallel to the exposure axis through the toe 
and shoulder breaking-points of the positive curves. Although great 
care was exercised in determining these points the accuracy of their 
readings can be disputed, because a rather large factor of inconsis- 
tency must be recognized due to the influence of directional developing 
effects upon the toe and shoulder. Furthermore, the breaking from 
the straight line naturally starts with very small angles of deviation, 
and for that reason, also, errors in determining the exact breaking- 
point are liable to occur. 

These lines form a border for the family of curves inside which all 
parts of the printed over-all curves are straight lines without distor- 
tion. A comparison of the length of the single straight-line parts 
within this territory, to the length of the whole curve used in the pro- 
duction of each over-all print, as determined, for example, by A and 
B, will clearly show the incorrectness of the conception frequently 
found that gamma is identical to contrast and is in all cases a means 
for determining the faithful photographic reproduction of the bright- 
ness range of an object. 

The law for ideal rendition of tone values by photographic means, 
according to which gamma negative is equal to the reciprocal of 
gamma positive, is applicable only in case gamma is identical to con- 
trast, which naturally is true only as long as one deals with straight- 
line recording and reproduction throughout. In processing motion 
pictures a large portion of any projectable over-all print, at least with 
the present technic, can not avoid being recorded in the curved toe 
section of the positive film. The correct equation, therefore, will 
stipulate that the over-all contrast or the product of negative gradient 
and positive gradient must equal unity. 

Let us now investigate how close these printed over-all curves 
satisfy this requirement. Considering the selected printed curve of 
laboratory No. 2 (Fig. 2) we have a printed over-all gamma of 1.25. 
Lines drawn parallel to the density axis through density 1.66 (A) and 
0.27 (B) will mark upon the exposure axis the exposure range related 
to the print. A line drawn parallel to the density axis through the 



Sept., 1935] 



PROCESSING CONDITIONS 



243 





Fio.3 




FiG.5 





Fio.4 




FIG 6 



FIGS. 1-6. Sensitometric curves for various printer lights obtained from 

the various laboratories (negative and printed over-all curves) : abscissas, 
exposures; ordinates, densities. 

Fig. Lab. Fig. Lab. 

11 45 

22 56 

33 67 



244 



H. MEYER 



[J. S. M. P. E. 



toe breaking-point of the printed over-all curve and crossing the ex- 
posure axis will determine the exposure range related to both the 
straight-line and curved toe portions. In this case it will be found that 
the exposure range related to the straight-line portion of the printed 
over-all curve approximates closely one half the total exposure range, 
which leaves the other half of that related to the curved portion. The 
average contrast of the curved part can be approximately determined 
by connecting B and the toe breaking-point with a straight line 
and reading its gamma value, which will be found in this case 
to be 0.80. Adding this value to 1.25 we obtain a value of 2.05, 
which, divided by 2, will give the figure 1.025 for the approximate 
over-all contrast, which is ideally close to unity. 

To avoid misunderstanding, it 
should be mentioned here that 
the method of interpolating over- 
all contrast as demonstrated in 
this example, can not be consid- 
ered scientifically correct or even 
practically applicable in all cases. 
It leads, however, to values suf- 
ficiently accurate for practical 
consideration in cases where the 
extension of the exposure range 
related to the curved section is 
not larger than that related to 
the straight-line portion. This 
method was employed here merely 

for the simple help it offers in illustrating how an over-all print with a 
gamma value considerably higher than unity can still render satis- 
factory contrast. 

The extension of the straight-line portion, in comparison to that of 
the curved portion, will naturally change in every print, depending 
upon the contrast and the density range of the negative, the printing 
light, the intensity range of the printer, and other variable factors in 
exposing and processing motion pictures. 

One will further notice when studying the over-all print graphs that 
the gamma values do not change, regardless of the printer light used. 
Small deviations noticeable in these graphs can be attributed to errors 
in density readings, unavoidable inaccuracy in determining the 
breaking-points of toe and shoulder of both the negative and the 




FIG. 7. Same as Figs. 1-6, for 
laboratory 8. 



Sept., 1935] PROCESSING CONDITIONS 245 

positive curves, and to distortions caused by directional developing 
effects. An automatic lowering of the contrast is encountered as 
soon as the densities of the over-all print fall below, or above, the 
area fenced by the various lines drawn through the points of toe and 
shoulder breaks. This, while undoubtedly known and recognized 
since sensitometric curves were introduced, is mentioned for the 
reason that the practical laboratory man usually thinks and expresses 
himself in terms of gamma, and, perhaps not fully aware of the differ- 
ence between contrast and gamma, believes quite often that what he 
notices as a lowering of contrast in a print, due to toe distortion, is 
caused by lowering of the gamma. 

The possible tonal distortions in a picture print, one of which is 
always present in the form of the positive toe distortion, can be 
studied from these curves as to their relative importance. The nega- 
tive toe distortion can be avoided in most types of photography pro- 
vided sufficient lighting is available, as modern negative film emul- 
sions are distinguished by a latitude or extension of their straight-line 
portion amply large enough to take care of the brightness range in 
almost any object. In this connection must be mentioned the de- 
sirability of having a large range of printing lights available which 
will permit an exposure through the highest negative densities, 
especially since gray-back types, with their additional over-all den- 
sity, are commonly used. This leads to the practical consideration 
not to exaggerate or generalize upon the merits of so-called low-key 
lighting. Only in cases where the brightness range of an object is 
greater than can be recorded within the straight-line portion of the 
negative film should the negative toe portion be used. Furthermore, 
this is preferable to using the shoulder portion, because the distortion 
caused by the latter will be doubled in the print by the unavoidable 
toe distortion of the positive film. In addition, distortion ot the 
shadow rendition in the print, as caused by portions of the negative 
being recorded in the negative toe section, is less noticeable to the eye. 

The distortion caused by the positive shoulder break is of less im- 
portance for the same reason and, besides, is practically avoided in 
current motion picture processing due to the fact that the highest 
densities in a projectable print seldom extend beyond the shoulder 
breaking-point of present positive emulsions. 

At this point it might be timely to apply some of these considera- 
tions to the two principal methods used in Hollywood laboratories for 
developing negative picture film, which are commonly known under the 



246 H. MEYER [J. S. M. P. E. 

names of "Test Method" and "Time and Temperature Method." It 
is not our intention to discuss the practical merits of either method or 
to express a final opinion as to which one is preferable in practical 
application. The laboratory, or studio, will decide this question 
primarily from the point of view of economy in time and labor, and 
guided by the desire of creating the best means to protect the all- 
important quality of original production negatives. 

Our consideration is limited to the purely theoretical question as to 
which of the two methods is more satisfactory as regards obtaining a 
negative that will, under prevailing positive processing conditions, 
result in an over-all print with an average over-all contrast close to 
unity. There can not be much doubt that in answer to this question 
the "test" method must be regarded as superior. 

In outdoor photography, for instance, maintenance of a constant 
lighting contrast is beyond the control of the cameraman. In addi- 
tion, most laboratories, since the introduction of developing ma- 
chines, keep their positive developing conditions practically constant 
as regards speed or time. Last, but not least, it should be remem- 
bered that the motion picture industry, up to the present time, is 
using positive film of one grade in contrast only, and does not utilize 
different grades, as does the paper print trade. Altogether, this seems 
to leave the only possibility of balancing inherent light contrasts to 
the negative processing. To exercise this method correctly one must, 
however, disregard the gamma and printing density of the negative. 
Just how far the "test" method is applicable, or preferable, in a prac- 
tical way is a question of a different nature, the discussion of which 
does not enter these considerations. 

A practical suggestion might be offered and utilized by the labora- 
tories using either method, which would call for a test slate similar to 
that represented by the black-and-white split background, and which 
would be exposed by the cameraman before the actual scene is taken. 
This would enable the laboratory to determine density range and 
contrast in conjunction with the negative solution curves, and could 
even be worked out without the direct aid of photography into some 
general method of measuring the values of highest and lowest bright- 
ness of such a test object under the given illumination by means of a 
photometer upon the set. The figures obtained could be checked 
against a standard and finally transposed into negative density con- 
trast for a given type under constant developing conditions. 

Furthermore, consideration was given to the possibility of avoiding 



Sept., 1935] PROCESSING CONDITIONS 247 

the use of the positive toe completely in processing the print so as to 
obtain a straight-line reproduction from a straight-line negative. 
This could be done by developing the print to an over-all gamma of 
unity and treating this print with a subtractive reducer to the point 
where the density of the toe portion is chemically dissolved. In order 
to attain the requisite projection density this print would either have 
to be correspondingly overexposed before development and reduction, 
or the normal exposed print would have to be re-intensified after 
reduction. 

Again it must be mentioned that it is not the purpose of this paper 
to deal with the practical application of this suggestion which un- 
doubtedly presents difficulties and disadvantages not easily over- 
come. It will be seen from the study of these tests that even the 
most perfect processing conditions at present will lead to results 
which still represent only a compromise to the ideal. Fortunately, 
the human eye, which is the final judge of our photographic efforts, is 
more tolerant than the critical ear with its objective selectivity. 

A study of this test material is recommended especially to the 
practical laboratory man, not so much for any novelty that it might 
contain as for the possibility that it offers to establish more definite 
knowledge of the requirements and limitations in the art of processing 
motion picture film. 

Sincere appreciation is hereby expressed to the major studios and 
laboratories whose courtesy and cooperation have permitted these 
tests to be conducted. 

REFERENCES 

JONES, L. A. : "On the Theory of Tone Reproduction with a Graphic Method 
for the Solution of Problems," /. Soc. Mot. Pict. Eng., XVI (May, 1931), No. 5, 
p. 568. 

LEAHY, W.: "Time-and-Temperature vs. the Test System for Development of 
Motion Picture Negatives," /. Soc. Mot. Pict. Eng., XVIII (May, 1932), No. 5, 
p. 649. 



NEW EMULSIONS FOR SPECIAL FIELDS IN MOTION 
PICTURE PHOTOGRAPHY* 

W. LEAHY** 

Summary. Three special film types, Infrared.Finopan, and Super pan Reversible, 
recently produced by the Agfa Ansco Corporation, are discussed regarding their photo- 
graphic characteristics and possible usefulness in special fields. Technical data as to 
speed, color-sensitivity, and physical properties are given for Infrared and Finopan, 
in comparison with those of Super pan negative film. 

The progress in photochemical research during recent years has 
resulted in the creation of highly perfected negative emulsions avail- 
able for general use in motion picture photography. The develop- 
ing motion picture technic has never ceased to demand continuous 
development of certain characteristics in these film types, and it can 
be readily seen today that the various special requests put before 
the film manufacturer will lead sooner or later, to an increased develop- 
ment of specialized film types. 

Any combination tool, while it is very convenient sometimes, will 
never give the performance of a line of special tools, and this also 
applies to a negative emulsion manufactured for all-around use. 

There are two principal reasons for the development of specialized 
emulsions, one of which is the trend of motion picture technic itself 
toward specialization, as can be seen, for instance, in the fact that 
certain types of exposures formerly made by the cameraman on the 
"set" are now placed in the hands of special departments. The 
second motive leading in this direction is the fact that the film manu- 
facturer, with his established knowledge of photochemical laws, feels 
more and more that despite all his technical skill and experience, 
these laws definitely limit further improvement of certain qualities 
of his products unless other qualities are sacrificed. To reach, for 
instance, the last possible step in general photographic speed, a prod- 
uct would have to be created that would disregard graininess, keep- 
ing quality, and contrast. To produce a film type of minimum 



* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Agfa Ansco Corp., Pacific Coast Technical Division, Hollywood, Calif. 



248 



NEW EMULSIONS 249 

graininess a reduction in speed and an increase in contrast will be 
unavoidable. 

It might, therefore, be only a question of a short time when the 
development of specialized film types having single outstanding 
characteristics will be the last possibility of satisfying the multiplying 
requests of the motion picture industry for further improvements 
in special film. 

In what follows, three special emulsions recently produced by the 
Agfa Ansco Corporation are discussed. 

(1) Infrared Film. This film type is used for recording long-dis- 
tance shots in outdoor scenes which are obscured by haze, for ob- 
taining special cloud and night effects in daylight, for aerial photog- 
raphy, and for medical and other scientific purposes. 

The general speed of Infrared is approximately one-half that of 
Superpan, that is, when both types are exposed without filters and 
developed to the same gamma. 




FIG. 1. Spectrogram of Infrared film. 

Infrared film, however, must be used with red filters, as it is sensi- 
tive to blue light rays like all silver bromide emulsions. It is not 
sensitive to green-yellow, which permits the use of relatively light 
red filters, as it is only necessary that these filters absorb blue. For 
this reason, also, the filter factors are practically the same for all 
blue-absorbing and red-transmitting filters which have approximately 
the same transmission factors within the visible range of the red end 
of the spectrum. All Wratten filters from monochrome No. 21 up to 
29-F fulfill this requirement and will be found to have equivalent 
exposure factors. Even filters as light as Wratten No. 12, Minus 
blue, and 15-G are suitable for most cases, although both transmit 
some ultraviolet in the wavelength range of 300 A. The filter fac- 
tor for Infrared in combination with these filter types, as found by 
practical tests and sensitometric comparison, is of the order of 10 to 
15. At standard motion picture camera speed a normal exposure of 
Infrared, using Wratten filter No. 25, will be obtained with a lens 
opening of 5.6. The use of deeper red filters is not recommended, 



250 



W. LEAHY 



[J. S. M. P. E. 




except for special scientific work, as they unnecessarily prolong the 
exposure due to their lower transmission factor without rendering 

better picture quality. 

Fig. 1 is a spectrogram of Infra- 
red indicating the color-sensitivity 
over the full range of the visible 
spectrum. 

Fig. 2 shows graphs of sensito- 
metric curves exposed on Infra- 
red film in an Eastman time- 
scale sensitometer, developed for 
different times in a regular mo- 
tion picture negative borax de- 
veloper. The gamma-time curve 
and the fog-density-time curve 
are also inserted in these graphs. 

Sensitometric curves on'lnfra- For comparison of relative con- 

redfilm. trast, similar sensitometric curves 

were made on Agfa Superpan and 

developed in the same developer, as shown in Fig. 3. It will be no- 
ticed in these that the gradation of Infrared film is considerably 
steeper than that of Superpan. Exposure of Infrared film through 
red filters naturally causes an 
increase in contrast, which was 
found to be approximately 7 per 
cent, referring to increase in 
gamma values. 

The sensitometric curves shown 
in Fig. 2 were developed using a 
green safelight, Agfa No. 103. 
Green filters permit the transmis- 
sion of infrared rays to some de- 
gree, but fog an infrared sensitive 
emulsion during an extended de- 
velopment. This is evidenced in 
the fog-density-time curve shown FIG atoo** curves on Agfa 
in Fig. 2, which marks the rapid Superpan. 

increase in fog density with ex- 
tended developing time. For normal developing time, however, it 
is permissible to use green lights with the ordinary precautions. 




Sept., 1935] 



NEW EMULSIONS 



251 



(2) Finopan Film. This film type is principally characterized by 
extremely fine grain, even excelling that of Infrared film. It might, 
therefore, be considered for use in photographic work wherein graini- 
ness and definition are of special importance. In general speed, Fino- 
pan is approximately 2 to 3 times slower than Superpan, the Weston 
rating being 8 for daylight and 4 for Mazda light. 




FIG. 4. Spectrogram of Finopan. 

Fig. 4 is a spectrogram of Finopan demonstrating the fully pan- 
chromatic response over the color range of the visible spectrum. 
Fig. 5 shows graphs of sensitometric curves exposed on Finopan 
in an Eastman time-scale sensitometer, developed for different times 
in a regular motion picture nega- 
tive borax developer. The 
gamma-time curve and the fog- 
density-time curve are also in- 
cluded in the graph. 

The contrast of Finopan is also 
higher than that of Superpan, as 
will be seen upon comparing the 
gamma readings on both film 
types for identical developing 
times. 

Special attention is called to 
the lack of fog, even for abnor- 
mally extended developing times, 
and also to the unusual length 
of the straight-line portion. 




FIG. 5. 



Sensitometric curves on 
Finopan. 



This film type, due to a superior 
grain quality, should be of interest 
in photographing negatives for background projection prints. Its 
natural contrast might be utilized in the production of titles and 
inserts. Finopan is also recommended for consideration in the proc- 
ess of making duplicate negatives, in which case, however, special 
handling is required during development to take care of the contrast. 



252 



W. LEAHY 



[J. S. M. P. E. 




Dupe prints have been successfully produced both by optical and 
contract printing, exposing the lavender print from the original nega- 
tive on light 21 and developing in a negative picture solution to a 
gamma of 0.70. The dupe negative was then made on Finopan, 
which was fully exposed and developed to a gamma of 1.0. The 
final print was very satisfactory, as regards grain iness and contrast, for 
the reason that by means of this procedure it is comparatively 
simple to register the full range of densities in both the lavender and 
the dupe negative upon the straight-line portion. 

A practical difficulty, however, of applying the method to the 
technic of making duplicates was recognized in the fact that the 
studios in many cases utilize stock library lavender 
prints for making dupe negatives, which naturally 
are developed to a normal positive gamma. 

(3) Reversible Superpan. During recent years 
a number of Hollywood motion picture studios 
have experimented with the problem of utilizing 
the reversible process in motion picture produc- 
tion. So far these attempts have failed mainly 
because of the limitations of this process itself. 
It seems to be somewhat early to announce or 
discuss reversible film types at a time when pro- 
fessional motion picture laboratories are not yet 
acquainted or experienced with processing rever- 
sible film. Therefore, it should be understood 
that mention of this film type is made merely with 
the intention of pointing to ward future possibilities, 
Super- which must be fully appreciated before beginning 
work toward perfecting methods of utilizing them. 
The advantages that this film seems to offer in connection with the 
reversible process include not only that of finest possible graininess 
but, perhaps more important, that of incomparably better registra- 
tion and definition due to circumventing the printing operation. 
Reversible Superpan is practically equal in general speed to regular 
Superpan negative, its Weston rating being 24 for daylight and 12 
with Tungsten light. 

In Fig. 6 is shown a single frame of a normal print made from a 
Supersensitive panchromatic negative in comparison with a reversed 
positive picture, photographed on reversible Superpan, using the 
same object and identical exposure. 




FIG. 6. Frame of 
normal print from 
supersensitive pan- 
chromatic negative, 
compared with re- 
versed positive on 
reversible 
pan. 



Sept., 1935] NEW EMULSIONS 253 

As mentioned in the introduction, the three film types discussed 
are not intended to serve for general purposes. For that reason 
their usefulness and, in consequence, their importance to the motion 
picture industry will be limited. They form, however, a definite step 
in the direction of specialization, which will make further develop- 
ments in response to the increasingly critical requirements and de- 
mands of motion picture technic. 

While additional film types will undoubtedly multiply his pro- 
duction problems and worries, the film manufacturer is sincerely 
willing to work in this direction, convinced that his attempts will be 
appreciated by the motion picture industry as a substantial help in 
solving their technical problems. 



TECHNICAL ASPECTS OF THE MOTION PICTURE* 
ALFRED N. GOLDSMITH** 

Summary. It is shown that the motion picture industry involves the artistic utili- 
zation of the products of research and development in the fields of mechanics, sound, 
heat, light, electricity, and chemistry. It is urged that the closest cooperation between 
the aesthetic and technical groups be maintained to the end that increasingly effective 
tools should be produced by the technicians and that these tools may be successfully 
employed by the artistic groups. 

Depending upon one's point of view, the motion picture in its 
present form may be any one of a number of things. It may be re- 
garded as an art, or as an industry, or a mode of entertainment, or as 
a means of instruction. If it be an art, like other arts, it will require 
appropriate tools. Considered as an industry producing a given 
product, it will need suitable machines. Considered as an entertain- 
ment or instructional agency, it will obviously demand some meth- 
ods of effective entry to the eye and ear of the audience or stu- 
dents, as the case may be. Analyzed from any angle, motion pictures 
are not an abstraction but a concrete reality and, as such, they re- 
quire the physical means to produce them and to make them effec- 
tive in their chosen function. 

Thus the technical aspects of the motion picture, contrary to the 
belief of some, are not annoying oddities or irrelevant intrusions; 
they are, instead, essential parts of an organic unity which is the 
production and reproduction of the sound motion picture. Why 
then object to them any more than one would object to the typewriter 
with which the "shooting script" is produced or the liquid powder 
which may be the base of a make-up ? A painter might as well detest 
easel, paint, canvas, and brushes ; an etcher, copper plates, engraving 
tools, and acid; or a sculptor, model-stand, clay, marble, chisels, and 
shaping tools. It is incongruous and unjust to fail to give proper at- 
tention to the technical aspects of the motion picture, and it involves 
the risk of missing great opportunities for progress in the art. Let us 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** New York, N. Y. 
254 



TECHNICAL ASPECTS OF MOTION PICTURES 255 

never forget that motion pictures are paintings in light and scrolls 
in sound. The mastery of light and the control of sound are at the 
very core of this great industry. 

Before considering these matters in somewhat greater detail, we 
may first pay a tribute of genuine respect and admiration to the au- 
thors, writers, directors, scenic artists, actors, and all those who con- 
tribute the essentially aesthetic, human, and creative impulse to 
pictures. Their work has held the people of the world enthralled for 
the last few decades and promises to hold the absorbed interest of 
humanity far into the future. May we, however, bespeak their con- 
tinued friendly consideration for that group of their fellow-workers 
who, striving behind the scenes and far less well known to the public 
and even to the industry, none the less give their best efforts to shap- 
ing and handling the exquisitely delicate and amazingly complicated 
tools of the field. I refer to the research workers, development en- 
gineers, cameramen, sound recordists, projectionists, and other groups 
who are in close contact with the daily problems of the physical por- 
tion of the industry. 

It is indeed hard for those without practical experience to under- 
stand the difficulties of research. Only those who have struggled 
with the obdurate obstinacy of nature in attempting to wrest from her 
one of her secrets ; only those who, after months or years of unremit- 
ting toil in the search for a better instrument to be used by the in- 
dustry, have been compelled to accept the bitter conclusion that they 
were on the wrong track and have been forced to start out once more 
in a new direction ; only those who have endured the unending strain 
of effort and expenditure of time and money involved in the dis- 
covery of improvements of one sort or another only those of such 
experience can fully appreciate what is involved in research and 
development as they are now carried forward. The odds are great, 
the risks appalling, and the rewards and recognition only too infre- 
quent. Yet along this hard path lies the real future of the industry 
so far as it depends upon the availability of those tools which will 
inspire and strengthen the artist, and please and hold the audience. 

To consider somewhat more fully the contacts between the motion 
picture and science in general, it may be useful to adopt for the mo- 
ment a rather old-fashioned classification of the physical and chemical 
sciences. It was customary in the past to divide science into the com- 
partments of mechanics, light, heat, sound, and electricity; and then 
to throw in chemistry for good measure. The distinctions between 



256 A. N. GOLDSMITH [J. S. M. P. E. 

these divisions have, in the light of our present-day broader knowl- 
edge, worn so thin as to be almost meaningless but, nevertheless, 
they may serve here as a convenient basis for the classification of 
some of the technical aspects of the motion picture. 

The domain of mechanics is drawn upon heavily by the motion 
picture field. A wide variety of the most diverse materials are used, 
and their mechanical properties are of importance. Among such 
materials are the film base, various metals and alloys used in camera, 
recorder, and projector construction, and a number of more special 
materials used as acoustic insulators or reflectors. Various mechani- 
cal movements are extensively used, both of the continuous and in- 
termittent varieties. Camera cranes and other devices required for 
special shots can be correctly designed for easy and reliable operation 
only if correct mechanical principles are utilized. At the other end 
of the range from such massive structures, we find that a study of the 
vibrations and mounting of the diaphragms or other elements used in 
recorders and loud speakers is needed for successful design. One 
could multiply these examples to a considerable extent without 
exhausting the contributions which mechanics can make in motion 
picture technology. 

So far as the field of sound is concerned, it is, of course, self-evident 
that a field which today depends so markedly for certain of its effects 
upon sound must lean heavily upon acoustic principles and practice. 
The transmission of sound, the absorption and reflection of sound, 
and the insulation of sound are all important elements in the design 
of studio sets, in microphone placement, in successful recording, in 
informative review room auditions, and in adequate theater repro- 
duction. As a matter of fact, the design of the microphones, re- 
corders, studios, theaters, and sound-heads and loud speakers of the 
reproducing equipment (which, in their totality, form an elaborate 
and necessarily consistently coordinated system) constitute one of the 
most difficult and severe examples of acoustic engineering so far en- 
countered. There are so many variables in recording and reproduc- 
ing sound for motion pictures, and some of these factors are under 
such imperfect control, that it is not astonishing that even now the 
industry is just beginning to attain the essential mastery of the cor- 
responding technic and equipment construction methods. 

The field of light or optics is older in its applications to the motion 
picture in view of the evolution of the present performances from the 
older silent picture. Accordingly, one might expect the technic of 



Sept., 1935] TECHNICAL ASPECTS OF MOTION PICTURES 257 

camera, lighting, and projection to be definitive if not final. Oddly 
enough, such is not the case, and there seems to be many points at 
which the nevertheless generally acceptable practice of today might 
yet be improved. The controlled placement, intensity, diffusion, 
and direction of sources of studio lighting (with the related matters of 
perfected design of reflectors, condensers, and diffusers) are suscep- 
tible of further development along lines of correct optics and electrical 
and mechanical engineering. Photometric and photographic mea- 
surements utilize optical principles broadly and are regarded as matters 
of further development. To the scientific student of light, the speci- 
fication and construction of color filters at this time are admirable in 
some respects, but standardization in this field will nevertheless re- 
quire a considerable amount of further study. It need hardly be 
added that the optics of camera lenses (both for ordinary and color 
photographic systems), of finders, of the lighting system of printers, 
of projector condensers and reflectors, and of projector objective 
lenses have been highly developed and are excellent examples of the 
contributions of the theory of light to the motion picture field. Fur- 
ther applications of the available principles and methods bid fair to 
improve current practice in these fields as well. 

The science of heat has contributed rather less to motion pictures 
than the preceding. While the control of the heat produced by vari- 
ous illuminants and the air conditioning of studios, laboratories, and 
theaters are examples of the applications of thermal theory, this 
field, on the whole, is not so directly related to motion picture prac- 
tice as some others. 

On the other hand, electrical theory and practice have made most 
substantial contributions to motion picture technology. Electrical 
generators and motors are extensively used throughout the industry. 
Portable power plants for "booster" illuminating installations are an 
interesting example of such an application. Camera motors, re- 
corder motors, and printer and projector drives are other examples. 
Rotary converters or dynamo tors are useful adjuncts. Illumination 
throughout all divisions of the industry is electrical, and is controlled 
by more or less advanced electrical methods. Switching comes into 
use in the studio and in connection with stage, house, and marquee 
and lobby lighting, where it has the indispensable function of render- 
ing the theater attractive to the public. More recently the technic 
of electron tubes as amplifiers and the like has found extensive and 
indispensable application in sound recording and reproduction, and 



258 A. N. GOLDSMITH 

it is not to be questioned that it will in due course find further appli- 
cation in the work of the cameramen and projectionists. So wide is 
the application of electricity to motion picture work that it can truly 
be said that one not versed in electrical circuit practice is at distinct 
disadvantage in the field at present. 

And last, but by no means least, the science of chemistry is widely 
applied in motion picture equipment and practice. The film itself 
indispensable forerunner of all that follows is strictly a complicated 
and beautifully controlled chemical product. The base, the emul- 
sion, the backing, the paper for wrapping the film, the developing, 
fixing, and hardening baths (and their maintenance) all make ex- 
treme demands upon the skill of the chemist. One might go into 
considerable detail on the numerous complicated problems which 
are here encountered. At the other end of the motion picture proc- 
esses, namely, in the arc-lamp chamber of the projector in the theater, 
one finds an interesting example of combined chemical control and 
spectrophotometric applications in the production of arc carbons 
which have just the right characteristics to give an abundance of 
properly tinted light upon the screen. Half-way between, the pro- 
duction of the coatings in the photo-cells of the sound-head and the 
production of the amplifier tubes, all involve many chemical problems, 
some of which are not yet wholly solved. 

It is believed that it will be agreed by those who candidly consider 
the foregoing comments that motion pictures are undoubtedly a 
fabricated technologic product, fashioned and controlled in skilled 
artistic and inspirational manner. For its continued preeminence 
and public acceptance, those charged with the responsibility for the 
direction of the motion picture field must encourage its technical as- 
pects and learn ever more fully to utilize them. They must respect 
and foster the engineering workers who are able to contribute to the 
art. And above all, there must be no conflict, open or hidden, be- 
tween artistry, craftsmanship, and technology for among these there 
is an essential underlying unity, the realization and encouragement of 
which is a fundamentally necessary condition for the continued 
growth and success of the industry. 



SOME TECHNICAL ASPECTS OF RECORDING MUSIC* 
R. H. TOWNSEND** 



Summary. A short account of some of the early difficulties of introducing sound 
commercially and artistically into the motion picture, followed by a description of the 
development of dubbing, substitution of voices, recording music, etc. Some of the 
problems of recording sound for long shots, short shots, etc., variation of distance 
between actor and microphone, are discussed, in addition to various methods employed 
in scoring musicals and voices for long-shot and short-shot quality. 

The entertainment value of any motion picture is a direct measure 
of its success at the box-office. Just what constitutes entertainment 
is a moot question, and there is no place in this paper for an academic 
or philosophic discussion on the subject. However, in making a 
picture there are several very definite factors that can and do con- 
tribute to the pleasing or disturbing effects that a picture will pro- 
duce upon an audience. 

These factors lie definitely within the realm of engineering as 
applied to picture making, and while the statement may sound 
strange, it is nevertheless true that a judicious correlation of engi- 
neering principles and psychological reactions become a vital necessity 
in producing any successful sound picture. 

We know, for instance, that a good story is most essential; that 
the cast should consist of individuals fully capable and desirous of 
portraying the characters in the story; that the photography should 
be adequate, and that the lines read by the cast should be so read that 
the audience can without effort believe them. 

Almost any good story will make a good picture, and it is believed 
that no good picture has been classed as poor entertainment because 
of improper lighting or distorted sound. These may detract from 
the over-all charm and intelligibility of a picture, but the picture 
may still be a good one in spite of photographic shortcomings and 
furnish entertainment for those who view it. 

Photography and sound have long been considered the technical 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Fox Film Corp., Hollywood, Calif. 

259 



260 R. H. TOWNSEND [J. S. M. P. E. 

phases of picture production, and the application of each of these 
arts is capable in no small measure of adding to or detracting from the 
general entertainment value of any production. 

Before the advent of sound, motion pictures had attained a high 
state of development, in the art of pantomime, and in extreme 
flexibility of action. An audience, figuratively speaking could 
be taken instantly to any part of the world, into the future, into the 
past, and back into the present again, all within the space of a few 
minutes. The shadowy individuals upon the screen could move 
about close to or far away from the camera without producing any 
conscious awareness upon the part of the observers that the subject 
had moved. In short, for a full and complete enjoyment of the 
picture the audience was called upon to exercise but one of its five 
senses, i. e., that of sight. 

One October night in New York City in 1927, however, something 
happened that projected the entire motion picture industry into a 
series of artistic, technical, and directorial convulsions, and which, 
in spite of a wide variety of sedatives and palliatives in the form of 
executive and engineering experimenting, still recur at altogether 
too frequent intervals. This was the night Warner Brothers' Jazz 
Singer electrified Broadway in more ways than one. 

Over night almost, the whole technical structure underlying the 
production of motion pictures experienced a tremendous shake-up. 
Any one who has had any connection with the industry during the 
past seven years has at least some idea of what has happened. Flimsy 
out-door platforms, known as stages, covered with cheap muslin 
screens, were replaced at a cost of millions of dollars with massive 
concrete structures, the walls of which were thick and solid enough 
to withstand artillery fire. Stars whose names had blazed from 
electric lights upon thousands of marquees all over the world suddenly 
faded into oblivion because the public was unable to reconcile queer 
sounding vocal renditions with beautiful profiles, voluptuous curves 
and manly figures. Somehow it didn't seem quite right to see upon 
the screen a real "he-man" with a gun in each hand galloping into 
the scene upon a fiery broncho and then hear him in a thin falsetto 
voice demand that the villain "unhand that there maiden." It was 
a real calamity for a producer to be faced with the realization that 
his glamorous star of the heretofore silent screen, when the story 
called for a song to her lover in the moonlight, had a vocal range of 
less than half an octave and most of her few tones just a little bit flat. 



Sept., 1935] TECHNICAL ASPECTS OF RECORDING 261 

Nothing in the line of lighting, photography, action, or beautiful 
scenery could overcome those handicaps. Something had to be 
done and out of that something has been developed a recording and 
re-recording technic that, even in its present state, is truly amazing. 

One of the first problems confronting producers after it was demon- 
strated by the first successful sound picture that sound was here to 
stay, was the disposal of the silent pictures already made and those 
in production. For the most part, pictures already completed were 
temporarily held up, and a print sent to a recording laboratory which 
up to that time had been devoted entirely to the production of 
phonograph records. A musical score for the entire production was 
composed and arranged by the laboratory staff personnel together 
with composers and arrangers employed by the picture producer. 
Recordings were made of the score and the picture released with a 
disk record of music accompanying each reel of picture. 

Later, as recording equipment became available to the picture 
studios, certain sequences in pictures were made with sound; dialog 
and sound effects were recorded at the time the particular scenes 
were photographed. The remainder of the picture carried a musical 
score with occasional incidental sound effects, and the entire pro- 
duction was released upon either film or disk. Most of the early 
productions were made with disk records. The installation of sound 
reproducing equipment in the theaters was meanwhile progressing, 
and more and more pictures were being made complete with sound 
until today practically everything is released only upon film. 

The fact that in early pictures only certain scenes were recorded 
with dialog, because other scenes had already been photographed 
and could be retaken only at great expense, brought into operation a 
system called "dubbing." The operations were essentially these: 
a sequence already photographed was projected upon a screen; the 
characters who originally appeared in that scene were assembled 
before the screen, and each at the proper times spoke his lines into 
a microphone, and the voices were recorded upon a recording ma- 
chine which was electrically synchronized with the projector. The 
resulting sound-track was then printed with the picture and the 
result was a sound picture. Sometimes the audience wondered why 
the lip motion of the characters upon the screen seemed to be ahead 
of or behind the voice, but for the most part the illusion was adequate 
to "get by" largely because of the novelty of the entertainment. 



262 R. H. TOWNSEND [J. S. M. P. E. 

The next step was the substitution of voices. If a character had 
a poor recording voice, why couldn't a good voice be used instead? 
It could, and was in many instances done. It was only a short time 
until a beautiful girl who couldn't carry a tune in a basket appeared 
upon the screen to be singing with all the assurance and delightfulness 
of a grand opera star. As many a star of the olden days was provided 
with a double for risky or dangerous stunts, so was an inarticulate 
star of the new regime provided with a vocal double whose tender 
tones added additional charm to the already popular screen per- 
sonality. At the present time, however, little or no vocal substitu- 
tion is done in pictures for the stars whose names grace the marquees. 

There is, however, a very considerable amount of dubbing being 
done at the present time by all studios. During the filming of a 
scene there may be reasons why a good recording of a voice is im- 
possible to obtain, and in such instances the scene is shot silent and 
the voice recorded later. The speaker watches the projected picture 
and speaks into a microphone, fitting his phrases to the lip motion of 
the character upon the screen. The sound-track is later printed with 
the picture. 

Making musical films presents a multitude of problems, each of 
which requires its own solution and not any of which is present in 
making an ordinary talking picture in which only dialog is to be 
recorded. In many instances, a musical picture is more or less 
spectacular inasmuch as it involves magnificent, lavish sets, hundreds 
of persons, and a great deal of action. The principals must, of 
course, have their share of close-ups, and almost invariably there 
are scenes involving the principals and a large chorus at the same 
time. 

To record an entire musical satisfactorily thus becomes a major 
problem, and requires untold ingenuity upon the part of the sound 
department. Take, for example, the filming of one scene of a typical 
musical. There is usually a prima donna, a leading man who sings, 
perhaps a trio or a quartette supporting the two, and a mixed chorus 
in the background. In order to accommodate such a large cast the 
set must of necessity be large. It is usually photographed with 
anywhere from three to six cameras, each of which uses a lens of 
different focal-length. Hundreds of incandescent lamps and arcs 
pour light into the set, and when everything else is ready, the un- 
fortunate sound man is faced with the task of putting a microphone, 
or a series of microphones, in places where he can secure a good sound 



Sept., 1935] TECHNICAL ASPECTS OF RECORDING 263 

pick-up and, at the same time, not cast shadows upon the scene or 
the actors. The chances of his accomplishing this are somewhat 
less than zero and, consequently, he has to resort to quite different 
means. 

On such a problem the sound man is severely handicapped. The 
cameraman can set his camera in any one of a dozen positions, close 
to or considerably removed from the object or objects to be photo- 
graphed, and by choosing any of half a dozen different lenses, he 
can attain the desired photographic result. The sound man, on 
the other hand, is forced to use a metal ear in the form of a micro- 
phone, and no matter how many microphones he may use, or try 
to use, he is able to accomplish only the effect of dividing a single 
ear into that many parts. Where it is possible to photograph a 
close-up and a long shot with two cameras simultaneously and secure 
excellent results in each case, a sound man can record either close-up 
quality or long-shot quality. He can not have both, because he has 
but a single recording channel with which to work. 

In order to overcome such a condition and to record successfully 
in spite of the very definite limitations imposed upon him due to this 
"distance factor," the sound engineer is compelled to resort to all 
sorts of manipulations. The fact that he has at his disposal a re- 
cording channel with certain predetermined electrical characteristics 
that remain constant does not alter or improve the situation con- 
tinually confronting him in his attempt to make a single sound record 
match a half dozen different camera shots of a single scene. 

It is comparatively easy in a broadcasting studio, where sound 
is of paramount importance, to secure the proper voice quality and 
balance, because there the microphone can be placed in an optimal 
position with respect to the speaker, singer, or other sources of sound. 
Upon a motion picture set, however, obtaining a satisfactory pick-up 
becomes an entirely different matter. In broadcasting, a micro- 
phone can be placed where it will do the most good, and usually that 
place is directly in front of the artist. In pictures such a placement 
is obviously impossible, and the microphone must be placed above 
and in front of the speaker. This immediately brings in at least 
two distinct complications. 

The first is the matter of shadows, for no matter what else happens 
one can not have shadows upon either the face or the clothes of the 
artist, nor on the walls or furnishings of the set. In the event that 
the subject stands still during the take, there is ordinarily little likeli- 



264 R. H. TOWNSEND [J. S. M. P. E. 

hood of shadow trouble; but if the action involves movement by the 
artist, then the difficulty of avoiding shadows is greatly increased 
because most of the set lighting is from overhead. 

The second problem confronting the sound man has to do with a 
change of voice quality as the microphone is moved with respect to 
the speaker. In real life we are accustomed to listen to a speaker 
from a position somewhat directly in front of him. In pictures we 
listen to him from a position above his head, since this is where the 
microphone is invariably placed. As long as the microphone is 
fairly close to the speaker it receives a large amount of direct sounds 
and very little of reflected or indirect sounds. As the distance be- 
tween the microphone and the speaker increases, the amount of 
direct sound picked up decreases and the amount of indirect sound 
reflected by the walls and floor of the set increases. This produces 
a very definite quality difference which is much more noticeable 
with a microphone pick-up than when listening with unaided ears. 
In the former case the reverberation appears to be much greater 
and if the microphone is moved too far away from the sound source, 
the reflected or indirect sounds increase to such an extent as to render 
the recording altogether unsatisfactory. 

Fortunately for the sound man, it is seldom in these days that a 
director will insist upon shooting a close-up camera and a long shot 
camera at the same time. It is usually only on large spectacular 
shots that more than one camera is used, and in such instances only 
mass sounds or crowd noises are recorded when the picture is photo- 
graphed. 

Assume a typical case: the proposed action, let us say, requires 
that the two principals be found sitting upon a bench in a garden. 
Romance is in the air, and under the soft moonlight the hero pours 
out his heart in a love song. He sings a verse and a chorus. She 
sings a chorus, then they sing a duet, and while they are singing a 
group of beautiful girls come into the picture, dance through a chorus, 
and the scene ends with a grand ensemble of all the voices and a 
full orchestra. 

In general, this is how such a scene would be made. Several days 
before the scene is to be photographed, the orchestra and singers are 
assembled in the recording room where all music is recorded. Several 
sound-tracks are made of the vocalists accompanied by the orchestra, 
and if there is to be a dance chorus, there is a recording of the or- 
chestra alone. The following day the sound-tracks are heard by 



Sept., 1935] TECHNICAL ASPECTS OF RECORDING 265 

those concerned and the one most suitable as to tempo, rendition, 
etc., is selected for use. 

A print of this recording is then taken to the set on the day when 
the picture is to be photographed, and used for playback purposes. 
The cast is assembled upon the set, and the record is played back 
to them as many times as necessary for rehearsal. When the routines 
are established and everything is in order, the scene is shot with 
cameras only while the cast go through the action to the recorded 
music from the reproducer, which is electrically interlocked with the 
cameras. When close-up shots of the principals are desired, the 
cameras are moved into position and the scene is repeated either 
wholly or in part, but each time to the same playback record. 

By this means any number of camera angles can be obtained. 
Later any combination of these shots can be intercut, and the final 
assembly will match, not only as to pitch and tempo, but the action 
and sound will be perfectly synchronized as well, since all the picture 
cuts are printed with the original sound-track. Of course, it is im- 
portant that during close-up shots the artist be careful to move his 
lips to conform to the words coming from the playback horns. In 
the event that either or all the principals can not sing it is easy to 
substitute singing voices of other artists during the original recording 
and the scene is shot with the screen star before the camera and the 
sound-track of the other voice on the playback machine. This 
method is usually known as "pre-scoring." 

There are many other ways by which these results may be ac- 
complished and it may be of interest to discuss some of them briefly. 
One way is to make a recording of an orchestra only, playing an 
arrangement in the form of an accompaniment for a song. At 
another time, perhaps weeks later, a singer goes into the recording 
room and makes a record of the words of the song while listening to 
the orchestral accompaniment in a pair of ear-phones. The two 
records upon separate films can then be re-recorded into a single 
track of voice and orchestra. The combined track may then be 
played back upon a set while the artist does the scene before the 
cameras. This picture may then be printed with the re-recorded 
sound-track to produce a composite film. This method has the 
advantage that it permits the balance between the voice and the 
orchestra to be altered at will during the re-recording process. For 
example, if an orchestra is shown in the picture and the singer moves 
about, perhaps to some distance from the orchestra, while the camera 



266 R. H. TOWNSEND [J. S. M. P. E. 

moves with him, the illusion is aided by dropping the orchestra level 
while still holding the voice at a normal level. 

Another method is to record a piano track of the selection to be 
used in the picture and reproduce this track upon the set at a volume 
level just loud enough for the artist to hear while he does his routine 
or song before the cameras and microphone. The picture and the 
piano track are then taken to the re-recording department where an 
orchestra is assembled. The picture is shown upon a screen in the 
recording room where a musical director can see it, and as he sees 
the picture he listens to the piano track over a pair of ear-phones. 
The reproducing machine and the picture projector are electrically 
interlocked; hence the two run in synchronism. 

After a few rehearsals a recording is made of the orchestra playing 
the necessary accompaniment while the director beats time for them 
to the tempo of the track he hears reproduced in the head-phones. 
Sometimes only the piano track is used and sometimes both the 
vocal track and the piano track are reproduced for the director. 
Of course, the microphone during this recording "hears" only the 
orchestra. At a later time the vocal track shot upon the set and the 
orchestra track are combined by re-recording and cut into the finished 
picture. 

Another variation in recording technic may be adopted when a 
singer and an orchestra are available in the recording room. It is 
always desirable to match long-shot sound with long-shot picture and 
close-up sound with close-up picture; but when a pre-scoring job is 
done, no one knows when and where the picture will be cut to these 
shots because no picture has yet been made. If now recordings 
are made of the singer and orchestra with close-up quality throughout, 
this sound will not match a long-shot picture when they shoot it, 
and the audiences will experience a feeling of dissatisfaction and will 
sense the unreality of the illusion, probably without knowing why. 
On the other hand, if the recording is made with long-shot quality, 
it will be equally bad when played with a close-up picture. To 
overcome this trouble and to provide for any amount of cutting later, 
any combination of three different methods may be used to record 
the original sound: 

(a) The sounds may be picked up by a microphone or micro- 
phones placed near the singer and orchestra, and called "close-up" 
quality. At the same time another microphone placed some dis- 
tance away from the source of sound feeds "long-shot" quality sound 



Sept., 1935] 



TECHNICAL ASPECTS OF RECORDING 



267 



to a second recording channel and a second recorder which is elec- 
trically interlocked with the one on the first channel. Either track 
may be used later upon the set as a playback while the picture is 
being photographed, and this reproduced sound recorded upon the 
set at the same time, but for cueing purposes only when the picture 
reaches the cutting room. 

The picture may then be cut in any desirable way, and when a 
long-shot scene is used, the corresponding portion of the original 
long-shot sound-track may be cut in. When a close-up picture is 
used, original close-up sound-track is used, and the finished picture 




___ 

FIG. 1. Schematic diagram of a circuit for producing artificial reverberation 



will then preserve and present to the audience at least a semblance 
of reality. 

(b) A modification of this procedure may be made use of as 
follows: Recordings are made simultaneously with long-shot and 
close-up microphones, and either track used for playback purposes 
while the picture is being photographed. However, instead of cut- 
ting in long-shot or close-up track into the finished picture, another 
procedure may be followed. The two tracks are reproduced simul- 
taneously by two high-quality reproducers and re-recorded into a 
single track. By watching the picture and by means of cue marks, 
the re-recording engineer can fade in or out either track at will, 



268 R. H. TOWNSEND 

thus accomplishing electrically what was, in the case of (a), done 
with a pair of shears, and often do a better and smoother job. 

(c) A third method of creating the desired illusion is to record a 
single track of the singer and orchestra with microphones so placed 
as to attain a satisfactory over-all balance with respect to each, and 
without regard for either close-up or long-shot quality. A print of 
the track thus made is then used for playback purposes, as in the two 
preceding illustrations. Whether this reproduced sound is or is 
not recorded upon the set during the filming is unimportant except 
for checking synchronization when the "rushes" are screened the 
following day. A print of the original track is then taken to the re- 
recording department, reproduced on a high-quality "dummy" and 
re-recorded. During this re-recording process, however, a little 
"doctoring" is done. Reference to Fig. 1 will serve to make clear 
what happens. 

The sound-track is reproduced from any one of several "dummies," 
fed into a preliminary equalizing amplifier and thence to a mixer or 
control panel in the usual way. By means of a simple switching ar- 
rangement this signal is divided just ahead of a mixer position (7), 
part of it going into the mixer pot and part of it being fed through an 
amplifying system and into a speaker unit. 

The speaker unit is located at one end of a highly reverberant room, 
or echo chamber one having hard plaster walls and ceiling and a 
wooden or concrete floor. The room need not be very large, the 
shape being more important than the actual cubical content. At 
some other point in the room is located a microphone which picks 
up the signal from the speaker. This signal is increased through a 
regular microphone amplifier and fed back through a trunk-line into 
another position (4) on the mixer panel. 

It is now possible to combine electrically a portion of the original 
signal from the dummy and another portion of the same signal after 
it has passed through a time-delay circuit. The time delay is con- 
trolled by the distance between the microphone and the speaker in the 
echo chamber. In addition to the time delay, this chamber, being 
highly reverberant, imparts a distinct quality change to the original 
signal. The engineer, by varying the relative amounts of each of 
these two signals coming into his mixer panel, is able to change the 
character of the resulting signal from the original quality with little or 
no reverberation to the other extreme where it may sound as though 
the original recording were made in the Grand Central Terminal. 



REPORT OF THE PROJECTION SCREEN BRIGHTNESS 

COMMITTEE* 

The standards of the Society contain, at present, a list of nine items 
under the heading "Recommended Practice." These items, while 
not so fundamentally important as the dimensional standards, are 
valuable as suggestions for the guidance of the industry. A recom- 
mendation of screen brightness should logically be included in this 
section. 

The newly organized Projection Screen Brightness Committee has 
been given the task of preparing a report upon the basis of which the 
Society can make such a recommendation. The job is not a new one 
in the history of motion pictures. It is, indeed, somewhat discourag- 
ing to discover that, in the past twenty years, no less than five pre- 
vious committees of our own and other societies have worked upon 
this specific problem. Most of these committees have attacked the 
problem by attempting to work as groups to gather data on existing 
conditions in the theater. 

The present Committee feels that it can best serve the Society, not 
by turning itself into a research body to obtain additional data on 
theater screen brightness, but rather by stimulating individual au- 
thors to prepare reports dealing with the most important phases of 
the subject. What the Committee proposes is that a symposium of 
papers shall be presented at a forthcoming meeting of the Society, 
and that these papers shall be followed by a final report of the Com- 
mittee which will summarize the whole situation and will make 
definite recommendations. 

SCREEN BRIGHTNESS SYMPOSIUM 

The following proposed program of the screen brightness sympo- 
sium is, of course, only tentative. Additions and changes will, no 
doubt, seem necessary as the work progresses. It is the hope of the 
Committee that our selected authors will be able to gather up the 
loose ends of the existing data and fill in the missing items to such an 
extent that definite recommendations can be made. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 

269 



270 SCREEN BRIGHTNESS COMMITTEE 

(1) A study of the literature bearing upon the question of screen 
brightness (covering all known relevant data of physiological optics 
and suggesting subjects for research necessary for a complete under- 
standing of the problem. This paper should consider the work of 
Helmholtz, Konig, Trolland, Nutting, Reeves, Feree and Rand, 
Luckiesh and Moss). 

(2) An analysis of the published results of theater and screen 
illumination measurements (covering the various papers on the sub- 
ject as, for instance, those of Dennington, Burrows, and the various 
Committee reports). 

(3) An analysis of release print characteristics (a statistical study 
of the highlight, shadow, and average transmission of scenes of re- 
lease print quality). 

(4) An experiment to determine the screen brightness requirements 
of the public (a statistical study of the desires of a typical audience 
the matter of individual taste, the influence of subtended visual 
angle, the influence of auditorium illumination) . 

(5) A note on laboratory screening room measurements (an account 
of the data obtained in release-print laboratories, with a discussion of 
the relation between theater and laboratory screen brightness). 

(6) Methods of measuring screen brightness (a paper to guide the 
Committee in its forthcoming recommendation of standard practice 
in measurement). 

(7) Projector characteristics, screen characteristics, and obtainable 
brightness (an attempt to combine the existing data on screen reflec- 
tion and source output in the form of tables and curves usable by 
the theater manager). 

(8) A report by the Committee (final recommendations of (a) 
screen brightness limits, (>) method and instruments of measure- 
ment, (c) best brightness as a function of screen size (visual angle) 
and auditorium illumination). 

PROJECTION SCREEN BRIGHTNESS COMMITTEE 

C. TUTTLE, Chairman 

G. A. CHAMBERS, Vice- Chair man 

J. O. AALBERG F. L. EICH G. F. RACKETT 

J. O. BAKER C. W. HANDLEY J. H. SPRAY 

B. BERG D. E. HYNDMAN A. T. WILLIAMS 

A. A. COOK W. F. LITTLE S. K. WOLF 

J. M. NICKOLAUS 



SYMPOSIUM ON NEW MOTION PICTURE APPARATUS* 

During the Spring Convention at Hollywood, Calif., May 20-24, 1935, a sympo- 
sium on new motion picture apparatus was held, in which various manufacturers of 
equipment described and demonstrated their new products and developments. Some 
of this equipment is described in the following pages; the remainder will be published 
in subsequent issues of the Journal. 

THREE NEW KODASCOPES 

N. B. GREEN** 

Model L 16-Mm. Kodascope. A successful 16-mm. projector must be attractive 
in appearance, easy to operate, simple to adjust, quiet and steady in projection, 
efficient in its optical system, cool to the touch, centrally controlled, mechanically 
satisfactory over long periods of time, and easily adaptable to all projection con- 
ditions or limitations. The design of the Model L Kodascope (Fig. 1) was studied 
mechanically, optically, and electrically, and then a casing designed to bring the 
whole into a unity without unpleasant projections, awkward corners, or conflict- 
ing angles. 

The controls for forward, reverse, and still-picture projection are all governed 
by one lever, which is mounted upon a single plate with the switches controlling 
the starting and stopping and turning the lamp on and off. In replacing a lamp 
the new lamp is properly positioned by turning a coin-slotted screw either in or 
out. Lenses, gates, condensers, and lamps are all easily removed for cleaning 
and as easily replaced. No tools are necessary. 

Four lenses and three lamps can be used on this projector, thus making possible 
a lens-lamp combination to suit a great variety of requirements. The lenses are: 
1-inch, //2, for use in close quarters; 2-inch, //1. 6, for average showings; 3-inch, 
//2, and 4-inch, //2. 5 for longer throws. Available lamps are: 400-watt, 115- 
volt T-10; 500-watt, 115-volt T-10; 750-watt, 115-volt T-12. All are of the bi- 
plane filament type, prefocus base. 

Every part of the frame and casing is of cast aluminum. Every shaft has 
double bearings, properly aligned and fitted. All shafts are ground within a 
tolerance of 0.0003 inch. All metal gears mesh with phenolic resin gears, which 
latter are mounted in metal hubs to guarantee against misalignment from warp- 
ing. The cams governing the intermittent are ground within a tolerance of 
0.0003 inch, and each intermittent is individually fitted. The result is a mecha- 
nism that will run, and has run, hundreds and thousands of hours without undue 
wear. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Eastman Kodak Co., Rochester, N. Y. 

271 



272 APPARATUS SYMPOSIUM [j. S. M. p. E. 

The cooling system is of great importance with the new high -wattage lamps. 
A centrifugal type of fan, driven by the projector motor, is mounted beneath the 
lamp, and delivers air to a housing (Fig. 2) around the lamp, designed to eliminate 
eddy currents and dead spots and flared above the lamp filament to allow for ex- 
pansion of the heated air. The cooling is so efficient that the lamp life in the 
Model L surpasses the rated life given by the manufacturers. 

The Optical System. Maximum screen illumination from the light-source has 
been obtained with this projector. The standard //1. 6 lens was designed espe- 
cially to provide flatness of field and maximum light transmission. With the 
single-blade shutter and this optical system, it has been possible to set a stand- 
ard of 210 screen lumens for the 750-watt lamp. 

Kodascope Eight, Model 80. With the omission of the reverse, the mechanism 
of the Kodascope Eight, Model 80 (Fig. 3), is simplified. As in the Model L, 
great care is taken in parts and assembly to the minutest detail; and the inter- 
mittent is blanked and formed to its final shape, and then copper-plated and 
shaved on its wearing surfaces only. It is then pack-hardened, and the copper 
plate is removed. This leaves all wearing surfaces hard, so as to withstand long 
service, but the body of the intermittent is soft and easily adjusted and fitted. 

The operation of the Kodascope Eight, Model 80, is very simple. The switch 
in the cord controls both lamp and motor, and has a receptacle for a table or floor- 
lamp. One lever controls still projection and rewinding. "Still" projection is 
accomplished by merely shifting the drive belt to an idler pulley. Rewinding is 
accomplished by merely threading the film back upon the empty reel and throwing 
the rewind lever. This is made possible by the use of smaller and lighter reels 
which carry a maximum of 200 feet of film. The belt tension is therefore light, 
and no undue strain is put upon the film during the showing of stills. The pat- 
ented elevating feature has a spring that compensates for the weight of the pro- 
jector and makes it easily operable either in raising or lowering. The same prin- 
ciples of cooling the lamp are used as in the Model L. Another patented feature 
is the arrangement made for adapting different lamps. Although the standard 
lamp is the 300-watt, coiled-coil lamp, the 300-watt, biplane filament can be in- 
stalled by loosening the socket clamp and resetting the socket. 

The optical system, designed for 8-mm film, compares favorably in efficiency 
with the best of 16-mm. projectors. The//!. 6 lens can be purchased as an accessory. 

Kodascope Eight, Model 40. The optical system of the Model 40 (Fig. 4) is 
remarkably efficient. The specially designed objective lens with the polished 
condenser and silvered glass reflector, together with the 200-watt lamp, give a 
screen illumination equal to that of the original Model 60. 

The mechanism is simple but rugged. The sprocket shafts and the intermittent 
shafts are hardened. The slower running shafts run in bearings drawn from the 
mechanism frame, but the faster running intermittent and shutter shafts run in 
bronze bearings. Oil reservoirs supply lubricant to the intermittent shuttle. The 
shutter has three blades of equal width. Fixed sprocket guides and strippers and an 
easy threading gate make loading extremely simple. Focusing is accomplished 
by turning the lens in or out. The cooling fan is mounted directly upon the motor 
shaft, and its housing is directly beneath the lamp housing. A simple baffle 
directs a blast of air toward the lamp but also allows a portion of the air to cool 
the lamp circuit resistor. 



Sept., 1935] 



APPARATUS SYMPOSIUM 



273 




274 



APPARATUS SYMPOSIUM 



[J. S. M. P. E. 





. 
w 



Sept., 1935] APPARATUS SYMPOSIUM 275 

A NEW SOUND READER AND FRAME VIEWER 

I. SERRURIER* 

A new machine added to the line of film editing equipment manufactured by 
this company consists of two parts, the machine head, Model SR V, and the com- 
bined amplifier and speaker, Model UR. The machine head (Fig. 5) consists 
of two curved film slides with a guide roller at each end of each slide. The film 
slide for the sound-film is equipped with an adjustable rail and adjustable guide 
rollers so that either standard 35-mm. film or split film may be used. An exciter 
lamp unit, the same as used on all Moviola sound reproducing equipment, is sup- 
ported above the small opening in the slide under the sound-track, and a caesium 
type of photoelectric cell is located in the compartment provided for it beneath 
this slide. The slide for the picture film has a frame-size aperture illuminated 
through an opal glass pane by a small electric lamp. A set of viewing lenses, 
the same as those used in all Moviola film viewing machines, is arranged over 
this illuminated aperture and can be hinged out of the way when threading the 
film over the slide and under the guide rollers. 

In Fig. 6 the machine head is shown with the viewing lenses swung away for 
threading. Two small switches are provided in front of the machine head, one 
for the exciter lamp and one for the viewing lamp. The machine head is con- 
nected to the amplifier and speaker cabinet by means of two shielded cables which 
may be readily detached, one for the 6-volt current supply to the exciter lamp and 
one for the photoelectric cell connection. 

The amplifier and speaker cabinet (Fig. 7) contain a 5-tube amplifier, a dy- 
namic speaker, and a transformer to supply the current for the exciter and view- 
ing lamps. These are the same as used in connection with all Moviola film edit- 
ing equipment. A knob for operating the volume control and a combined switch 
and red pilot light for the power supply to the amplifier are located on the 
front of the amplifier cabinet. 

To operate the sound reader it should be placed between two rewinders, pref- 
erably double rewinders, and the film or films should be drawn through it by 
means of the rewinders. It may be used to advantage in connection with a two- 
sprocket synchronizing machine, such as shown in Fig. 7, and should be useful in 
the cutting room for quickly locating words or sound effects to be added or elimi- 
nated. 

A JERK-ABSORBING DEVICE FOR FILM VIEWING AND SOUND REPRODUCING 
MACHINES WITH FILM ON REELS 

Moviola film editing machines are started, stopped, and reversed quite fre- 
quently, and the adjustment of the friction devices upon the take-up spindles is 
often neglected, with the result that some film may become unwound from a reel. 
When the machine is started and the slack is taken up the film must suddenly 
start the reel in motion, and a jerk results. To eliminate damaging the film from 
this cause, rollers have been added upon hinged and spring-held arms in such 

*Moviola Co., Hollywood, Calif. 



276 



APPARATUS SYMPOSIUM 



[j. s. M. P. E. 




Sept., 1935] 



APPARATUS SYMPOSIUM 



277 




FIG. 7. Complete set-up, with sound reader placed between two 
rewinders, showing amplifier and loud speaker cabinet. 




FIG. 8. 



Moviola Model UDS, equipped with jerk- 
absorbing device. 



278 APPARATUS SYMPOSIUM [J. S. M. P. E. 

manner that the roller gives when the film becomes taut, and the jerk on the film 
is thus eliminated. Fig. 8 shows a Moviola Model UDS equipped with these jerk- 
absorbing devices for forward and backward take-up devices of the picture head 
as well as of the sound-head. 



ARC SUPPLY GENERATOR FOR USE WITH SUPREX CARBONS 

W. K. HARTMAN* 

For a long time a demand has existed among the smaller theaters for a source 
of white light similar to that obtained from the d-c., high-intensity arcs and at a 
price that the average theater owner could pay. This demand seems to have been 
met by the new d-c., high-intensity arcs with non-rotating positive carbons. 

These carbons, commonly referred to as Suprex carbons, have unusual charac- 
teristics and are unlike any carbons previously used on d-c. arcs. It was imme- 
diately apparent that the existing conversion equipment to transform alternating 
to direct current was entirely unsuited to the new carbons. The 75-volt, multiple- 
arc set with an external ballast resistor in series with each arc, even with a changed 
ballast, resulted in a highly inefficient installation and unsatisfactory char- 
acteristics. 

The new carbons have a voltage drop of about 35 volts across the arc, as com- 
pared with 50 to 55 volts for the older low-intensity reflector type of arc. Fur- 
thermore, because of the snow-white light obtained, it is extremely important 
that a non-fluctuating, non-pulsating source of direct current be available. With 
normal current, the supply voltage should have a slightly drooping characteristic, 
and it is important that the striking current be kept down to avoid destruction 
of the crater and resultant poor light while the crater is being reformed. To ful- 
fill the exacting demands of the new carbons, this company has developed a new 
type of motor-generator set radically different from the older and more conven- 
tional type of multiple-arc or series arc sets that have been used for many years. 

The motor-generator set is built as two units, direct connected by a flexible 
coupling and mounted upon a rigid structural steel bed-plate. 

The motor is of the conventional design but the generator really consists of two 
independent generators in the same housing. These two generators are entirely 
independent electrically, and furnish direct current to two Suprex carbon arcs, 
each generator having in its own circuit a rheostat for independently adjusting 
its output and a voltmeter so connected that the voltage in either circuit may be 
read at will. 

In designing this machine, account has been taken of the mechanical require- 
ments of quietness of operation, rigidity, and accessibility. Mechanical quietness 
is obtained by careful dynamic balance of the armature, by use of a coupling with- 
out pins, requiring no lubrication, and with no parts to work loose and cause 
noise. The rotating element floats freely in sleeve bearings with end-play in 
either direction, and prevents any noise due to end-thrust. The four-bearing 
construction admits of ready accessibility and compactness. 

* Century Electric Co., Los Angeles, Calif. 



Sept., 1935] APPARATUS SYMPOSIUM 279 

In the electrical design of the machine precautions are also taken for quiet 
operation, because in the smaller and moderate-size theaters, the motor-gener- 
ator set is frequently mounted' adjacent to the projection room and any consid- 
erable noise would be objectionable. Magnetic noise is minimized by specially 
proportioning the windings of the motor and the generator with skewed armature 
slots and with specially shaped pole-tips. 

By virtue of the fact that there is an independent generator winding for each 
arc, it was possible so to proportion the windings that the voltage characteristics 
particularly suit the Suprex carbon; with relatively flat, slightly drooping 
characteristics in the normal operating range, and with a low striking-current to 
avoid the explosive effect at the crater, which would occur with higher amperage. 
Each generator operates independently of the other, and consequently is not 
affected by the load on the other (excepting minute voltage changes occasioned 
by changes in speed due to the load), so the change-over from one arc to two arcs 
is made without any disturbance in either electrical circuit. 

The motor is designed so that there are no excessive speed changes during the 
short-time peak load at change-over. The set has a speed of 1770 rpm. with one 
arc burning, and approximately 1745 rpm. with two arcs burning, operating on a 
60-cycle polyphase circuit. 

This design permits the elimination of all external resistances ; and by avoiding 
resistor losses the efficiency is kept rather high, being approximately 61 per cent 
with one arc burning at 50 amperes or 71 per cent with two arcs burning at 100 
amperes. The efficiency is somewhat lower with one arc in use because of the 
field losses occuring in the generator end that is not in use but is generating a 
voltage. 

This unusual motor generator set, which is sold under the trade name Actodector, 
has been developed after extensive field and laboratory tests and was placed on 
the market early in 1935, since which time many installations have been made in 
all parts of the United States. 



A PROFESSIONAL 16-MM. PROJECTOR WITH INTERMITTENT 

SPROCKET 

H. A. DEVRY* 

When the use of 16-mm. equipment began to be extended to sales and adver- 
tising activities of large firms for small group showings, and the possibility of the 
use of 16-mm. equipment in small-town theaters became apparent, we surveyed 
the field and decided to make an equipment particularly adapted, if possible, to 
more or less continuous service, and in this connection decided upon the use of an 
intermittent sprocket instead of the customary claw movement, in order to in- 
crease the life of the film and make its use most profitable. 

The 16-mm. equipment is now going through the same refinements and im- 
provements through which the 35-mm. projector went. All early motion picture 
machines before a standard 35-mm. film was evolved, and before the existence of 

*H. A. Devry, Inc., Chicago, 111. 



280 APPARATUS SYMPOSIUM 

the Society of Motion Picture Engineers, were of the claw type and, consequently, 
for exhibition purposes short-lived. The last of the claw type 35-mm. machines 
made and sold in this country for theatrical use was the Selig Polyscope, which 
engaged four perforations of the film at a time. This model was discontinued 
about 1910, and was replaced in the theaters by intermittent sprocket projectors, 
which rapidly found great favor among the exhibitors because of the longer life 
of the film made possible by using the sprocket instead of the claw movement. 
At that time rental films were scarce; most exhibitors bought their films outright, 
and traded or sold them after they had served their purpose. 

The regular Geneva movement, of course, is too slow and unsuitable for 16-mm. 
projectors, so we decided upon a modification of this, with an 8-point star- or 
pin-wheel and a larger and heavier cam. 

Special machinery had to be developed to cut a cam with a movement of about 
six to one or better, so that it would be practically noiseless. The star-wheel is 
revolved one-eighth of a revolution for each revolution of the cam, and is gener- 
ated to mesh into the cam in the same manner as a silent chain meshes into a 
sprocket, or as one gear meshes into another, so as to keep constant contact, as in 
the Geneva movement, and eliminate as nearly as possible all noise and wear. 

Casual tests indicate that the life of the film operating upon an intermittent 
sprocket driven by such a star and cam movement will be many times that of 
a film operated by a claw movement, to say nothing of the annoyance, incon- 
venience, and interruptions in showing film not in perfect condition when operated 
with the usual claw movement. 



EUGENE AUGUSTIN LAUSTE* 

In any future anthology of the inventors and inventions that have 
combined to make possible the modern sound picture, the work of 
Eugene Augustin Lauste, who died in Montclair, N. J., on June 26th, 
will inevitably receive important consideration. In terms of time, his 
experiments may be said to date before what have been described as 
the ''dawn days" of the motion picture and to extend down to the 
present time. 

A slight, shy Frenchman, Mr. Lauste lived at 12 Howard Street, 
Bloomfield, until his last illness. He was born in the Montmartre 
district of Paris on January 17, 1857. Surviving him are his wife, Mrs. 
Melaine Lauste, a son, Emile Lauste of London, and two stepsons, 
Clement LeRoy and Harry E. LeRoy, both of Bloomfield. 

The work of Mr. Lauste, starting in the eighties, represents a link 
between the group that later conducted epochal experiments in sound 
transmission and reproduction and the group that laid the ground- 
work for the future cinema by their experiments in step photography 
based upon the optical phenomenon of "persistence of vision." His 
inventions had to do with the fundamentals of both the silent and the 
sound motion picture. In 1889, while he was with Thomas A. Edison 
at Orange, N. J., he shared with W. K. L. Dickson in many of the 
experiments relative to the development of the kinetoscope and the 
kinetograph. In 1894, he became associated with Major Woodville 
Latham, and while in his employ he designed and constructed one of 
the earliest film projectors and cameras, the Eidoloscope. He was 
first to record sound and scene simultaneously upon the same film. 
His patent application filed in Great Britain on August 11, 1906, dis- 
closes most of the processes fundamental to the modern sound picture, 
except amplification. From 1906 to 1910 he devoted his efforts to 
attaining adequate results in sound recording and reproduction. Then 
one day in London, after he had recorded upon film a brief passage 
from a French gramaphone record by means of his first mechanism of 
the string galvanometer type, and was dubious of success, his amaze- 

* For a detailed description of some of Mr. Lauste's work, see J. Soc. Mot. 
Pict Eng. XVII (Oct., 1931), No. 4, p. 632. 

281 



282 EUGENE AUGUSTIN LAUSTE [J. S. M. P. E. 

ment and delight knew no bounds when he heard in his ear-phones 
very distinctly the words "J'entends tres bien maint. ..." From 
this point on his progress was rapid. He devised several types of 
recording apparatus using one and two non-magnetic wires in a mag- 
netic field, several of which are still in existence and illustrate his 
ingenuity as well as his fine mechanical ability. 

All present methods, however, of recording and reproducing sound 
for talking motion pictures rely for their practicability upon the 
vacuum-tube amplifier. Without amplification by that device, 




Eugene Augustin Lauste. 

the feeble current from the microphone would be unable to produce in 
the recording devices effects of sufficient magnitude. Without amplifi- 
cation the reproduced current could not operate high-power loud speak- 
ers. Given the possibility of distortionless magnification of a sound- 
bearing current, and also the general advanced state of the communi- 
cation arts to which the vacuum-tube itself has been no mean con- 
tributor, and there became possible some of the things for which 
Lauste struggled prematurely. 

He came to America for a short visit in 1911 with the idea of in- 
teresting capital. While in this country he photographed a short- 



Sept., 1935] EUGENE AUGUSTIN LAUSTE 283 

length picture, recording sound and scene upon the same film. This 
has been described as the first sound picture to be taken in the United 
States. He was attempting to devise an amplifier for his sound-films 
when lack of capital and the outbreak of the World War halted his 
experiments. He was still seeking financial aid in America when there 
came the sudden flood of sound-film patents. 

As for Lauste, himself, but for one happening, it might have been 
necessary to write finish to his career in the tragic fashion that has 
marked the conclusion of the histories of so many inventors, great 
and small. Seeking out the genesis of sound pictures in a painstaking 
effort to assemble an authoritative and historic record of the develop- 
ment of the new art, Bell Telephone Laboratories was impressed by 
the evidence of Lauste's pioneer work in this line. He was sought out 
in 1929 and retained as a member of its technical staff. He was 
assigned two tasks: one to assemble such of his own apparatus as 
could be located, and by replacements to reconstruct the entire system 
that he had built in the early 1900's; and the other to assist the patent 
department by his knowledge of the prior art. His contributions 
to the development of sound pictures may have failed to bring him 
the wealth for which he hoped or the success that is measured by 
money, but at least they sufficed to bring him a measure of the recog- 
nition, comfort, and security during his last years, that has more 
often than not been denied inventors whose success depended upon 
progress in allied fields. 

The Smithsonian Institution has accepted the gift of Lauste's 
historic apparatus. Accompanying this exhibit is a very complete 
record of photographs and manuscripts covering its authenticity and 
indicating which parts are restorations along the lines of the original 
design. 

Eugene A. Lauste was elected an Honorary Member of the Society 
of Motion Picture Engineers in October 4, 1931. He received his 
scroll in person while attending the banquet for the pioneers of the 
industry held at Swampscott, Mass., on October 7, 1931. 



FALL, 1935, CONVENTION 

WASHINGTON, D. C. 

WARDMAN PARK HOTEL 

OCTOBER 21-24, INCLUSIVE 

Headquarters 

The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations and Convention facilities are assured. Registration 
will begin at 9 A.M., Monday, October 21st. A special suite will be provided for 
the ladies. Rates for S. M. P. E. delegates, European plan, will be as follows: 

One person, room and bath $3 . 00 

Two persons, double bed and bath 5 . 00 

Two persons, twin beds and bath 5. 00 

Rates for connecting parlors 5 . 00 

A modern fire-proof garage is located on the Hotel property, and a special 75 
cents per day rate has been arranged for. 

Technical Sessions 

An attractive program of technical papers and presentations is being arranged 
by the Papers Committee. Sessions will be held in the Little Theater of the Hotel, 
off the west lobby, as follows: Monday to Thursday mornings, inclusive; and 
Monday, Tuesday, and Thursday afternoons. 

Film Programs 

Exhibitions of newly released motion picture features and short subjects will be 
held in the Little Theater on Monday and Tuesday evenings. Passes to various 
motion picture theaters in Washington will be available to the members register- 
ing for the duration of the convention. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held in the east 
lobby of the Hotel, to which all manufacturers of equipment are invited to con- 
tribute. The apparatus to be exhibited must either be new or contain new fea- 
tures of interest from a technical point of view. Information concerning the 
exhibit and reservations for space should be made in writing to the Chairman of 
the Exhibits Committee, Mr. O. F. Neu, addressed to the General Office of the 
Society at the Hotel Pennsylvania, New York, N. Y. No charge will be made for 
space. 
284 



FALL, 1935, CONVENTION 285 

Informal Get-Together Luncheon 

The usual luncheon will be held at noon on October 21st in the Continental Room 
of the Hotel. An address of welcome will be delivered by the Honorable Sol 
Bloom, member of Congress from New York. Other speakers will be announced 
later. 

Semi- Annual Banquet 

The semi-annual banquet of the Society will be held in the Continental Room 
of the Hotel on Wednesday October 23rd at 7 : 30 P.M. Addresses will be delivered 
by eminent members of the industry followed by dancing and entertainment. 
The presentation of the scroll of honorary membership to Thomas Armat, of 
Washington, D. C., awarded last May at Hollywood, will be made, and, in addi- 
tion, the recipients of the Journal Award and the Progress Medal of the Society 
will be announced and the presentations made. 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House, Library of Congress, Depart- 
ment of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of 
Justice, Archives Building; and other institutions such as the National Academy 
of Sciences, the Smithsonian Institution, George Washington University, Wash- 
ington Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many high- 
ways radiating from Washington. 

Recreation 

The Wardman Park Hotel management is arranging for golfing privileges for 
S. M. P. E. delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 

PROGRAM 

Monday, Oct. 21st 

9 : 30 A.M. Registration 
10 : 00 A.M. Society business 

Technical papers program 
12:30 P.M. Informal get-together luncheon 

2:00 P.M. Technical papers program 

8:00 P.M. Exhibition of newly released motion pictures 



286 FALL, 1935, CONVENTION 

Tuesday, Oct. 22nd 

10: 00 A.M. Technical papers program 
2: 00 P.M. Technical papers program 
8:00 P.M. Exhibition of newly released motion pictures 

Wednesday, Oct. 23rd 

10: 00 A.M. Technical papers program 

12:30 P.M. Free afternoon, for recreation or special trips and visits 
7 : 30 P.M. Semi-annual banquet 

Thursday, Oct. 24th 

10: 00 A.M. Technical papers program 
2: 00 P.M. Technical papers program 
6:00 P.M. Adjournment of the Convention 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV October, 1935 Number 4 



CONTENTS 

Page 

The Technical Aspects of the High-Fidelity Reproducer 

E. D. COOK 289 

Non-Theatrical Projection R. F. MITCHELL 314 

Studio Acoustics M. RETTINGER 331 

The Argentometer an Apparatus for Testing for Silver in a 

Fixing Bath W. J. WEYERTS AND K. C. D. HICKMAN 335 

Report of the Projection Practice Committee Projection Room 

Planning 341 

Report of the Sound Committee 353 

Apparatus Symposium : 

Ozaphane Film and the Cinelux Projector. .A. M. CHEFTEL 358 

A High-Speed Camera C. T. BURKE 360 

The Wall Motion Picture Camera H. GRIFFIN 363 

Fall, 1935, Convention at Washington, D. C 367 

Society Announcements 370 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



THE TECHNICAL ASPECTS OF THE HIGH-FIDELITY 
REPRODUCER* 



E. D. COOK** 



Summary. One of the essential requirements of a high-fidelity talking motion 
picture performance is a satisfactory reproducer. The most important limitation 
to such reproduction at this time, and one that is seriously handicapping those who 
desire to provide better recordings, is the general lack of sufficiently constant film speed 
in reproduction. This prevents full use of the advances made in recording machines 
and technic. In certain kinds of subject material, the performance is ruined for a 
critical audience by this limitation. 

For various reasons, little has been published on this problem. This paper de- 
scribes the technical details of the Photophone high-fidelity reproducer. Measure- 
ments of its mechanical constants, together with an elementary analysis of the under- 
lying theory of operation are given. 

From the layman's point of view, the talking motion picture has 
been regarded as a remarkable achievement. However, the engineer 
has realized that it was a combination of well-known ideas and that 
the real achievement lay in its improvement. It has always lacked 
realism. The range of frequency actually covered was quite re- 
stricted, wave-shape distortions were prevalent, the volume range 
available was relatively small, and what was probably worse was the 
lack of sufficiently constant motion in both recording and reproduction. 

During the early development, it was difficult to estimate how 
much of the latter defect was separately chargeable against the 
individual processes, but very soon better recorders were made 
available and laboratory reproducers were built upon similar prin- 
ciples. It was then quite simple to demonstrate that even the best 
commercial reproducers left much to be desired in this respect, if 
realism were sought. The limitation, imposed upon reproducers in 
common use by the lack of sufficient constancy of film speed, has 
been overcome in a large measure by the high-fidelity sound attach- 
ment shown in Fig. 1. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RCA Manufacturing Co., Camden, N. J. 

289 



290 E. D. COOK [J. S. M. P. E. 

In producing a device designed to make possible a higher standard 
of theater performance, there have been other problems such as 
those due to optical deficiencies, noises due to mechanical vibration, 
uneven and limited response of loud speakers, and in some cases 
even the amplifiers were not above reproach. Many seemingly 
trivial details related to the major problems had to be considered. 
That this was the case may be shown by a few examples. If the 
frequency range is to be widened materially in reproduction, the 




FIG. 1. The high-fidelity sound attachment (PS-24). 

general problem of ground-noise becomes more serious. The con- 
tributory factors have to be examined and more complete solutions 
provided than have been the case in the past. It might seem to be 
desirable to decrease the width of the reproducer aperture in order 
to decrease the high-frequency losses, yet this may not be continued 
to such an extent that the signal-to-ground noise ratio is likewise 
decreased to any considerable degree unless a more-than-offsetting 
gain in this ratio may be achieved in some other way, for example, 
upon the film itself, and it is evident that a definite limit exists in 
any such direction of attack upon the problem. 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 291 

Another detail is to be found in the photoelectric cell, its output 
coupling device, and associated amplifier. It is desirable that the 
amplifier be located remote from the projector, and in order to do 
this, a photoelectric cell coupling transformer and a relatively low 
impedance coupling line have been employed on Photophone equip- 
ment. With the extended frequency range, it was considered neces- 
sary to provide greater shielding against any inductive disturbances 
that might affect this circuit. In the attachment shown, the main 
metal casting has been used to assist in the shielding, and the photo- 
electric cell is located directly behind the drum so that the leads to 
the photoelectric cell transformer are not only quite well protected 
from outside disturbances but very short. This choice of location 
for the photoelectric cell has made threading a simple matter. 

Another problem is found in compensating the commercial repro- 
ducing amplifier. This can not be treated in the same manner as 
laboratory or demonstration equipment where the entire process of 
recording and reproduction is under more complete control. The 
products of various producers require varying amounts of correction 
for flat response at the higher frequencies. 

These questions, and many others, are vital in developing high- 
quality reproducing equipment. A discussion of them might provide 
the subject matter for an interesting paper, but since many of them 
are well known and some are adequately treated elsewhere, it was 
felt more appropriate at this time to consider the means employed 
in the PS-24 sound attachment to eliminate the specific and, in 
general, more serious problem of film speed variation. Although 
this always has been a major problem, for various reasons it has 
never been adequately treated in the literature. Only the phases 
of particular interest will be treated in this discussion. 

In general, reproducers of the drum type encounter more difficulties 
in regard to low-frequency speed variations, whereas those of the gate 
type are adversely affected more by the higher-frequency variations 
such as those due to sprocket holes, etc. This is not intended to mean 
that both types of disturbances are not found in machines of either 
type, for such a statement would not be correct. However, it is 
true that the more constant one desires to make the film speed, the 
more he is forced to use the drum design. In the reproducer to be 
described, high-frequency speed variations, such as those due to 
sprocket hole disturbances, are automatically reduced far below the 
requirements. This is a necessary consequence of the design of the 



292 E. D. COOK [j. S. M. P. E. 

mechanism, as will be evident from the theoretical analysis to be 
given later. Since variations of film speed can be impulsive as well 
as periodic, it is important to consider the design from the general 
standpoint. 

In order to judge any design from its filtering characteristics, it 
is essential to have some knowledge of the permissible velocity varia- 
tion and how this is affected by various conditions. Some general 
knowledge is available from listening tests, using records of different 
frequencies. It is known that if reflections are permitted from the 
walls of the room, the effect of speed irregularities is much more 
noticeable. Furthermore, the main reproduction frequency, as 
well as the frequency of variation and the general sound level, 
influence the permissible speed variation, and the results obtained 
from abrupt speed changes differ from those obtained from gradual 
ones, even though the total change is finally the same. The prob- 
lem is further complicated by the varying kinds and the complex 
character of the sounds that must be treated. If the somewhat 
contradictory evidence given by any two observers is to be trusted, 
the sensitivity to speed variation depends, to some extent at least, 
upon the individual and, no doubt, his reactions are influenced by 
his varying physical condition. Experience has shown that there 
are persons who are almost totally insensitive to the deficiencies of 
a device that would turn a grand piano into a Hawaiian guitar. Yet 
it is possible to assume some sort of a normal observer and average 
listening conditions, regardless of how mythical these may seem to be. 

No doubt considerable work has been done on this problem, but, 
so far at least, the published reports are few. Furthermore, it ap- 
pears that more comprehensive work has been done upon the differen- 
tial pitch sensitivity of the ear alone than upon the more general 
problem that exists when the reverberant effects of the room are 
included. 

The problem discussed by Shower and Biddulph 1 was the differen- 
tial pitch sensitivity of the ear alone (without room effects). For 
this reason, they employed telephone receivers and concerned them- 
selves primarily with single-frequency sounds. As should be ex- 
pected, their results for maximum permissible percentage frequency 
variation were higher than are permissible under the conditions of 
normal listening. Some of the general conclusions to be drawn from 
their results may be stated since they appear to be equally true when 
the reverberant conditions of the room are included. It was found 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 293 

that the rate of variation of frequency for which the ear is most 
sensitive is relatively low, corresponding to only several cycles per 
second. Furthermore, it appears that at moderately higher sound 
levels, the percentage of permissible frequency variation is lower 
than at the very low levels, while the higher "carrier" or main 
frequencies must be freer from variation than the lower ones. The 
manner in which the frequency variation takes place is likewise 
important, because, as might be expected, the ear is more sensitive 
to abrupt changes than to the smoother changes even though the 
periodicity of a cyclic change is made the same in both cases. 
However, the ratio of the sensitivity for abrupt changes to that for 
gradual changes of frequency decreases as the carrier frequency is 
raised above a few hundred cycles. 

It is possible to discuss this problem from the point of view of the 
amplitude of the distortion terms produced by the lack of sufficient 
speed constancy. Such a treatment has been given by Lautenschla- 
ger, 2 who concluded that the square-root of the sum of the squares 
of the amplitudes of the distortion terms should not exceed a value 
of approximately 7 per cent, if serious and objectionable distortions 
were to be avoided. 

Under certain conditions, even a frequency change of 0.1 per cent 
would be large enough to be noticeable; hence it would be desirable 
to keep as near to such a value as possible. There are too many 
projectors today having percentages greater than 1.0, with the result 
that in many theaters the reproduction is almost intolerable for 
music having sustained tones. However, it must be realized that 
the boundary between good and bad performance is not sharply 
denned, and perfectly constant speed is not attainable. Under 
average conditions, it is probable that a 0.2 per cent frequency 
change over the important range would be an acceptable value that 
would not be too difficult to attain. 

No attempt will be made to discuss the abstract theory of the 
fundamental phenomena involved here, but it may be noted in 
passing that the general subject matter is closely allied to that of 
frequency and phase modulation. The solutions of those problems 
have been available for many years, having been discussed by 
numerous writers, notably Carson, 3 Van der Pol, 4 and more recently 
by Roder, 5 who gives a rather complete bibliography. 

The effect of speed variation in a reproducer is to alter the funda- 
mental output and produce response at numerous side-tones lying 



294 E. D. COOK [J. s. M. p. E. 

above and below the fundamental reproduction frequency that might 
have been expected from the record. These side-tones correspond 
to the side-tones that exist in the better known case of radio trans- 
mission, and are spaced from the "carrier" frequency by multiples 
of the "wow" frequency. The amplitude of the various components 
is Bessel functions whose arguments are functions of what Van der 
Pol has called "the index of frequency modulation," or what may 
more crudely be called the amount of "wow." The significant thing 
about speed variation is that a whole new series of tones may be 
introduced into the reproduction which, besides having no counterpart 
in the original sound as recorded, may be totally dissonant with these 
sounds. 

As the frequency range of the reproducer is extended, this problem 
becomes more and more serious, and since the theater-going public 
is gradually coming to the point of view of purchasing a quality 
performance, the future can only force the quality standards toward 
better reproduction. 

The following analysis,* made by the author some years ago, 
proceeds from a point of view slightly different from that of similar 
analyses that have come to his attention. Let the ordinate of the 
wave recorded upon the film be defined by y at any phase position 0. 

y = A sin (1) 

Then the voltage developed across a pure resistance load for an 
ideal photoelectric cell may be represented by e when the angular 
displacement or phase of the record is <f> as it is moved past the re- 
producing aperture. 

e = E sin (2) 

* Analyses of this effect have been made by Lautenschlager 2 and Belar 7 , both 
of whom proceeded from the standpoint of a fixed amplitude, d, for the sinusoidally 
varying part of the film drum displacement, x. Their results, for a frequency 
having a wavelength X upon the record, are given in somewhat different form, 



(IT) si 



In the analysis given in this paper, the angular velocity of the film is assumed 
to be a constant, o> , with a superposed sinusoidal velocity of amplitude, Ob, vary- 
ing at a frequency of /3. 

There is no difference in the final result obtained from either relation provided 
only that the proper variables are correctly inserted in the expressions. 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 295 

If the angular velocity of the film is d$/dt = (co & cos 00* the 
phase at any time / is : 



or 

e = sin u t + sin j3t (4) 



for which the Jacobi-Bessel 4 6 expansion is: 

sin(co + 0)* (5) 



As the ''wow" is allowed to be greater due either to poor design or 
to some circumstance in the film, the amplitude of the fundamental 
or carrier frequency is decreased, and the side-tones extend further 
and further from the carrier frequency and may finally exceed the 
fundamental in amplitude. The circumstance to be regretted is 
really the dissonant beating effects with other and desired tones 
present in the record. Unfortunately, while the case of simple 
sinusoidal variation of film speed is serious, it is not the most serious 
case nor is it the only one met in practice. The abrupt variation is 
much more probable, and, as has been shown, the ear is more sensitive 
to such variations. 

Before discussing the theoretical aspects of the mechanical filtering 
in the high-fidelity (PS- 24) sound attachment, it is desirable to 
examine briefly the practical design features.* Some idea of the 
disposition of parts as well as the path followed by the film may be 
obtained from Fig. 2. 

In the design, it was necessary to consider carefully many such 
factors as the driving motor; the gears; the various effects of vibra- 
tion; induction pick-up; the path of the film on either side of, and 
adjacent to, the sound translation point; the required mechanical 
impedance of the film-moving mechanism at the translation point; 
the guiding; and isolation from the shocks from the take-up 
magazine. 

In so far as the driving motor was concerned, it was desirable to 
store as much kinetic energy as possible in the rotating element and, 
for that reason, a high-speed motor was used. It seemed preferable 

* The mechanical design of this projector attachment was the work of Messrs. 
F. J. Loomis and E. W. Reynolds. 8 



296 



E. D. COOK 



[J. S. M. p. E. 



to employ only a single-phase supply on a-c. circuits. In order to 
minimize the effects of shocks and line-voltage variation and to cause 
the readjustment to such conditions to take place over longer inter- 
vals of time, the induction motor was chosen for a-c. circuits. In 
special cases, three-phase induction motors have been used. In the 
case of d-c. circuits, additional inertia in the form of a flywheel was 
added to improve the speed constancy of the main driver. 




FIG. 2. The film path of the high-fidelity sound attachment. 

After leaving the last sprocket in the picture head, the film enters 
the sound attachment by passing, in a relatively free loop, to the 
combination guide and pressure roller which holds it in contact 
with the reproducing drum. The motion of the film upon the drum 
is controlled by the rotary stabilizer* a device that is fastened to 
the drum shaft, and which will be described later. These parts are 
shown separately in Fig. 3. 

* This device was proposed by Mr. C. R. Hanna and the present model was 
designed by Mr. E. W. Reynolds. 8 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 297 

As the drum rotates, the film is moved past the reproducing point, 
maintaining its contact with the drum over some considerable part 
of the circumference beyond this point by virtue of the film tension 
established by the main pulling sprocket. Since the film tension is 
a matter of only an ounce or so, the film leaves the drum in a rela- 
tively free loop, and engages the pulling sprocket in such a manner 
that the directions of the film upon leaving the drum and upon 
entering the pulling sprocket are nearly at right angles to one another. 
This feature has an important bearing upon the filtering action for the 
film motion while the film is in contact with the drum. The best 
method would have been to use a {/-shaped film loop between these 
points, but as that would have presented design difficulties, the pres- 




FIG. 3. Rotary stabilizer, film drum, and pressure roller system for sound 

attachment. 

ent method was chosen. The action may readily be visualized by 
holding a piece of film in the hand so that the described directions 
are obtained, and moving one end in a direction perpendicular to 
that of the other end. The motion imparted to the other end is 
considerably reduced. It is the existence of the film loops together 
with the relatively high mechanical impedance acting upon the drum 
shaft that so definitely eliminates sprocket -hole modulation in this 
machine. 

A second sprocket follows the pulling sprocket and is used to aid 
in isolating the reactions from the take-up magazine, which, because 
of their possible severity, might otherwise be transmitted through 
the film loops to the drum. 

The function of the rotary stabilizer is to provide an inherently 
damped positive reactance to cooperate with the negative reactance 
of the film loops in forming a mechanical filter. This device consists 



298 E. D. COOK [J. S. M. P. E. 

of a very light shell which is fastened to the shaft and within which 
a heavy flywheel is mounted upon a ball bearing in such a manner 
that it is concentric with the axis of the shell and may revolve as 
freely as possible. The clearance between the shell and the flywheel 
is purposely made very small, and the entire available space inside 
the shell is filled with oil. 

In operation, the inner flywheel revolves at film drum speed and 
if the film motion is not uniform, the motion of the film drum, and 
consequently the outer shell of the stabilizer is altered, while that of 
the inner flywheel is supposedly unaffected. There is a reaction, 
therefore, between the two elements, and a resultant dissipation of 
energy in the viscous medium that corresponds to a pure mechanical 





FIG. 4. A disassembled view of the rotary stabilizer. 

resistance. It is to be noted that this resistance is not operative in 
the case of steady motions, and therefore the steady film tension is 
not increased by employing it. The component parts of the rotary 
stabilizer are shown in Fig. 4. 

The mechanical filtering is accomplished largely by the coopera- 
tion between this device and the elastance of the film loops. Since 
this elastance depends upon the film tension, it is necessary to keep 
the tension quite low. Therefore, mechanical resistance (frictions), 
which would tend to demand more torque from the film-pulling 
sprocket, must be eliminated. To aid in accomplishing this, the 
pressure roller and the drum shaft are mounted upon ball bearings. 

It was apparent from the theoretical analysis of the stabilizer 
that in order to attain the desired filtering using a strictly elastance- 
resistance combination (without any inertia connected rigidly to the 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 299 

drum shaft), it would be necessary to provide a much greater viscous 
resistance than is used at present, because the elastance of the film 
loops is relatively fixed. This would mean either impossibly small 
clearances or increased dimensions, and either alternative would 
demand a heavier inner flywheel to prevent this member from follow- 
ing the velocity variations of the shell. Any variation of velocity 
of the flywheel due to a corresponding variation in the shell violates 
the assumption of an inertia-free film drum system because of the 
tightness with which these two systems are coupled under such 
conditions. 

It was evident upon theoretical grounds that improved filtering 
action was attainable in the higher-frequency region if some addi- 
tional inertia were allowed upon the film drum shaft. However, 
this inertia could not be too large in proportion to that of the inner 
flywheel of the stabilizer if oscillations are to be successfully damped. 
To establish the minimum permissible moment of inertia for the outer 
shell, various amounts of inertia were added to the film drum shaft. 

Several things were evident at once: first, with the possible film 
loop elastance available, it was found that the minimum of addi- 
tional inertia capable of keeping the speed variation of the drum 
within reasonable bounds was somewhere between 20 to 30 ounce- 
inches squared (in weight units); and, second, with such low-loss 
mechanical devices, at least some minimum of damping resistance 
was absolutely essential to prevent incessant speed variation. The 
damping obviously would increase with the inertia of the shell of 
the stabilizer and the stiffness of the film loops ; that is, if the energy 
stored during oscillation were increased, the means of wasting this en- 
ergy would likewise have to be increased. The exact relation between 
these constants will be deduced later in connection with the necessary 
conditions for critical damping. 

In the endeavor to obtain the theoretically indicated relation 
between the various constants to approach the critical damping 
condition as nearly as practicable, a certain size for the container 
shell was indicated even with the lightest possible shell design. This 
automatically provided several times the minimum required drum 
shaft inertia, and was therefore very useful in improving the filter- 
ing in the frequency range above approximately 3 cycles per second. 

Since the damping resistance is derived from the relative motion 
between the shell and the flange surface of the inner flywheel, it is 
essential that even the smallest significant change of velocity in the 



300 



E. D. COOK 



fj. S. M. p. E. 



drum shaft should find its counterpart between the component parts 
of the rotary stabilizer. It is therefore quite evident that the bearing 
friction of the inner flywheel must be as small as possible if the 
device is to function properly, and the static friction at this point 
should be no greater than the running friction, if possible. There- 
fore, ball bearings are indicated for this purpose. 

In the freely oscillating state, any variation of velocity of the film 
drum shaft corresponding to a definite change of kinetic energy 

stored in the drum system will 
be reflected as a change of po- 
tential energy stored in the 
elastance of the film loops. In 
this interchange of energy, as 
has already been indicated, the 
enclosing shell of the rotary 
stabilizer should be the only 
seat of kinetic energy change; 
the flywheel inside the device 
should have its velocity altered 
as little as possible if proper 
damping is to be obtained, or 
this oscillatory energy is to be 
wasted rapidly enough so that 
the film-moving system will re- 
turn to its non-oscillatory state 
as quickly as possible. Ob- 
viously, if the inner flywheel is 
too light, its velocity will be 
influenced too much by velocity 
changes of the enclosing shell, 

and hence the alternating difference of velocity between the shell and 
the flywheel will be decreased, with the attendant circumstance that 
too little of the oscillatory energy will be wasted in the viscous fric- 
tion between the surfaces of the two parts. Thus, there must be some 
minimum ratio between the inertia of the inside flywheel and that of 
the enclosing shell with its connected film drum and shaft, before any 
given damping condition can be attained. It is quite evident that 
the magnitude of the resistance coefficient required for any definite 
degree of damping in this device, although dependent upon inertia 
and elastance, is not determined as simply as for the case of the ele- 




FIG. 5. A device for measuring film 
elastance. 



Oct., 1935] 



THE HIGH-FIDELITY REPRODUCER 



301 



mentary circuit of an elastance, an inertia, and a resistance. 

It is desirable to show how the constants of the mechanical circuit 
were determined and their general order of magnitude before dis- 
cussing the theory of the mechanical operation of the reproducer. 

Attempts to determine the elastance of the film loops, as referred 
to the film drum by measurements made directly upon the projector, 
did not give as consistent results as might be desired, so the special 
device shown in Fig. 5 was built. The film path in this device was 
an exact duplicate of that used in the reproducer, and the frictions 




z 3 4 

RNGULftR DISPLACEMENT OF DRUM 

FROM 
EQUILIBRIUM POSITION 



FIG. 6. The force-displacement curve of film loops referred 
to film drum. 



involved in the essential bearings were reduced to a minimum. The 
measurements themselves consisted in applying a known torque to 
the drum by means of a weight, and noting the resultant angular 
deflection. Naturally, variations of the torque-displacement curve 
were found with various lengths of film between the two sprockets 
on either side of the film drum, but since the average condition is 
of greater importance at this time, only the curve for this case is 
shown in Fig. 6. This curve will naturally vary somewhat for dif- 
ferent machines in actual practice but such variations are not serious. 
It is seen that the torque -displacement curve is anything but linear 
and in some ways this may be advantageous. The complete theory 



302 



E. D. COOK 



[J. S. M. p. E. 



for large oscillations would have to consider a variable elastance; but 
fortunately, in operation, the oscillations are restricted to relatively 

small angular displacements, 
and, therefore, under normal 
conditions the elastance pre- 
sented to the alternating ve- 
locities is approximately con- 
stant. This allows the solution 
to proceed with acceptable ac- 
curacy along the conventional 
lines well-known in the general 
theory of oscillatory systems. 
For example, the variation of 
film elastance would be about 
10 per cent at a "wow" fre- 
quency of 2 cycles per second 
unless the total frequency 
change exceeded 0.3 per cent, 
which would be an appreciable 
speed variation even at a carrier 
frequency as low as 100 cycles. 
This discussion indicates that 
the effective film stiffness can be 
evaluated upon the basis of small displacements. Hence, the torque 
required to produce such a displacement should be written as : 

dT = dd 

where is the increment of film stiffness. Therefore, the film stiffness 
(Fig. 7) is derived from the torque-displacement curve (Fig. 6) by 
evaluating the slope at each position, or: 

- 

In one projector the film loops were required to transmit a force 
of approximately 36 grams to the drum during operation. In this 
case it will be seen from Fig. 7 that the incremental film stiffness was 
approximately 2950 gram-centimeters per radian. 

In determining the moment of inertia of the various parts involved 
in the mechanical filter circuit used to attain a sufficiently steady 
speed of the film upon the drum, use was made of the well-known 
properties of the torsional pendulum. A disk whose moment of 



FILM STIFFNESS (GM. CM. PER RADIAN) 

o I ! I ! 1 1 1 1 1 1 1 I I I I 1 I 1 I 1 












1 




























1 












/ 














1 












/ 














/ 














/ 












/ 














/ 














/ 














/ 














/ 












/ 














/ 














/ 












/ 














z 














"""^ 





























FORCE (SMS.) AT RADIUS OF DRUM 

FIG. 7. Incremental stiffness of 
loops referred to drum. 



film 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 303 

inertia, I s , was accurately known by calculation, was suspended 
horizontally by a very thin steel wire fastened to a rigid support. 
The period of oscillation, P s , was then carefully measured. The 
part whose moment of inertia was to be found was suspended from 
the disk so that its axis of rotation coincided with that of the tor- 
sional pendulum, and the new period of oscillation was determined. 
The moment of inertia, /, of the part to be measured, was deter- 
minable from the relation: 

I / D \ 9 I 

(7) 

where P is the period of oscillation of the combination of the unknown 
and the standard. Since the theory of this device may be found in 
any standard work on mechanics, it need not be given here. 

The moments of inertia found for the essential parts of the test 
reproducer are given in Table I. The accuracy of the results is 
quite sufficient for the present purposes. 

TABLE I 

Approximate Moment 
Approximate of Inertia 

Weight (Weight Units) 

(Ounces) (Ounce-Inches Squared) 

Outer shell 15.0 99 

Inner flywheel 78.5 462 

Film drum and shaft 13 . 25 1.5 

The final constant, necessary for calculating the performance of 
the rotary stabilizer, is the damping resistance due to the viscous 
oil film. A true mechanical resistance is one that requires a torque 
or force directly proportional to velocity, to overcome the friction. 
Viscous friction obeys this law. The constant is therefore deter- 
minable directly from the torque-speed curve of the device, and, like 
incremental film stiffness, is the derivative of the curve at the point of 
operation. 

In order to produce an anti-hunting device, the torque of the 
resistance must oppose the resistance applied. Therefore, the re- 
sistance is defined as: 



K> 

Equation (8) would be equally useful for those cases where the speed- 
torque curve should not happen to be a straight line, provided only 
that the oscillations were restricted to small amplitudes. This must 



304 E. D. COOK [j. s. M. p. E. 

be true in the present case in order to satisfy the conditions imposed 
upon a reproducer of this type. 

In the particular device involved here, the speed-torque curve is 
a straight line. This is fortunate from an analytical standpoint, 
because it makes the equations easier to deal with, and in practice 
the existence of constant coefficients prevents the generation of sub- 
harmonics of an applied sinusoidal force. For design purposes, the 
speed-torque curve can be calculated with sufficient accuracy for all 
practical purposes, from the following relation: 



where /JL = viscosity of the fluid 

I = axial length of outer cylindrical surface of inner flywheel upon which 

the torque is developed 

a = radius of inner surface of enclosing shell 

b = radius of outer surface of inner flywheel 

ft-5 = angular velocity of shell in radians per second 

fii = angular velocity of flywheel in radians per second. 

In this equation, the effects of the ends and edges of the cylindrical 
surfaces have been neglected, but this is not serious because the 
stabilizer has been designed to derive a minimum torque from these 
parts. 

The variation of the viscosity factor, /z, with temperature of the 
fluid was at first thought to be a serious handicap to the device. 
Several years of commercial use, however, have shown that the fear 
was unwarranted, for although the variation does exist, the necessity 
for maintaining critical damping, or any other given degree of damp- 
ing, has been found to be commercially unnecessary. It is essential 
only to damp the oscillations with reasonable rapidity. 

Several authors have attempted to discuss the variation of ju by 
empirical relations. Notable among such attempts is the work of 
Slotte, 9 who gives the following approximation : 

M = M o(l + /Sfl-if (10) 

where t\ and ft are constants for a given fluid and t is the temperature 
in degrees centigrade above the reference point for which the vis- 
cosity is HQ. 

Although the calculated torque-speed relation is useful for design 
purposes, the device must obey certain test limits in production; 
hence the mechanism shown in Fig. 8 was constructed to measure 
the developed torque. It is essentially a speed-changing gear-box, 
so arranged that the output shaft rotates always in one direction. 
The device is driven by a synchronous motor, and the speed changes 



Oct., 1935] 



THE HIGH-FIDELITY REPRODUCER 



305 



are made by shifting several external gears. Thus, the exact speed 
can be known as accurately as desired. 

The torque-speed curve of the stabilizer as measured on this device, 
is shown in Fig. 9. It is to be noted that this curve must vary some- 
what between samples because it is not economical to hold the manu- 
facturing tolerances too close, 
and commercially the results 
attained would not warrant this 
refinement. An average value 
of the resistance obtained in 
this manner would be approxi- 
mately 126.5 gram-centimeters 
per radian per second. 

The operation of the system as 
a mechanical filter can be readily 
appreciated by transferring the 
problem into electrical termin- 
ology regarding the velocity 
analogous to the current and the 
torque analogous to the voltage. 
It will be instructive to trace the 
circuit analogue for the present 
case, assuming that the more 
serious causes of speed variation 
reside in either of the sprockets 
on either side of the film drum. 
This assumption is entirely in 
accord with the facts. 

It must be noted at the outset that the mechanical system con- 
nected to either of these sprockets is very ponderous and stiff, and a 
relatively great amount of energy would be required to drive the 
projector from the sprockets, even assuming that the various parts 
would stand the strain. This merely means that if the source of 
disturbance is to be regarded as a generator, it possesses a very high 
internal impedance; which is merely another way of saying that the 
generator should be regarded as a constant-current device. A more 
detailed examination of the various kinds of disturbances will reveal 
the essential correctness of this view. Furthermore, since the elas- 
tances of the film loops upon either side of the film drum are to be 
considered jointly, the sprockets upon either side of the drum may 




FIG. 8. A device for measuring the 
damping resistance of the rotary 
stabilizer. 



306 



E. D. COOK 



[J. S. M. p. E. 



likewise be regarded as one generator having 'a current wave of 
complex shape. In the analysis, it will be sufficient to discuss the 
effect of a single sine wave, or, in the limit, the effect of an instantane- 
ous change of speed. 

It is noted that if the pulling sprocket is angularly displaced 
6 radians in time, /, and if the film is assumed absolutely inelastic 
for the moment so that all the motion is transferred to the film 
drum, the total displacement of the drum is less than that of the 
sprocket by the ratio of the diameter of the sprocket to the diameter 
of the drum. In this case, the ratio is approximately 0.578. More- 
over, if a given force is applied at the sprocket, still assuming a rigid 



r-124 5 CM.CM PER RfcDIfcN MR SICOND 

TBMP. Z4*C 
&ULF CKtST A OIL. 



R.P.M OF SHELL RELATIVE TO FLYWHEEL 



FIG. 9. The speed-torque curve of the rotary stabilizer. 

film connection between it and the drum, the force applied to the 
drum will be equal to it; but, because of the difference between the 
radii, the torque on the drum will exceed that applied to the sprocket 
by the ratio of the drum diameter to the sprocket diameter. It is 
obvious, therefore, that some form of transformer action exists 
between the two parts so that the velocities are decreased and the 
torques increased as they affect the drum. However, this ideal 
transformer likewise passes the steady motions. If the alternating 
effects are the only characteristics of interest, this action may without 
error be represented by an ideal, mutually coupled transformer. 
Fortunately, the case fulfills the requirements of the principle of 
superposition closely enough so that the direct-current action may 
be neglected in the present analysis, and hence no further time will 
be spent upon this added complication. In passing, it may be noted 
that the principle of superposition assumes linear equations. 



Oct., 1935] 



THE HIGH-FIDELITY REPRODUCER 



307 



Assuming that the character of the various circuit elements is 
known, the circuit diagram of the analogue may be readily traced; 
for, if a change of velocity is impressed upon the pulling sprocket, 
not all this change instantaneously reaches the film drum even after 
allowing for proper transformer action. Some is momentarily lost 
in the elastance of the film loops. Hence, it is seen that the loops 
correspond to a condenser connected in parallel with the film drum 
circuit. That portion of velocity change that affects the film drum 
encounters the inertia of the drum, the drum shaft, and the shell of 
the rotary stabilizer before passing on to the enclosed elements. It 
has been assumed that the friction and inertia of the film-guiding 
system, as well as the bearings of the drum shaft, are a second order 



* 



FIG. 10. The circuit diagram of the electric analogue 
of the film-moving system of the sound attachment. 

effect, an assumption that is warranted in practice. Furthermore, 
it is readily seen that any change of velocity of the film drum does 
not instantaneously reach the inner flywheel of the stabilizer, but 
some is lost in the viscous resistance of the oil film, through which 
the transfer to the flywheel is made. This loss is transformed into 
heat and is not regained. Thus, the viscous resistance is connected 
in parallel with the inertia of the inner flywheel in the analogue. 
The circuit diagram is shown in Fig. 10. 

where 5 = sprocket generator 

Zi = high internal impedance of generator 

i = impressed alternating velocity 

N p = primary "turns" of an ideal transformer (sprocket diameter) 

N s = secondary "turns" of an ideal transformer (drum diameter) 

a = ratio of secondary current to primary current = ( -* J = f - J = 

0.578 

elastance of film loops measured on film drum (impedance = Z) 
inertia of drum, shaft, and stabilizer shell 
inertia of inner flyWheel of stabilizer 

mechanical resistance due to viscous action as measured on film drum 
impedance of r and L in parallel 
impedance of / and Z in series. 



then 



(ID 



308 



E. D. COOK 



[J. S. M.P. E. 



or: 



a(r + pL) 



(12) 



\_pHLC + p\l + L)Cr +pL 
In order to give the complete solution for the expression, the type 
of stimulus and the roots of the denominator must be known. In 
any complicated numerical case, these may be determined by Dande- 
lin's 10 method. For the steady-state case, assuming that the applied 
velocity is a sine wave, the above relation is readily reducible to: 



d 



+ (coL) 2 



- o> 2 [/ 



(13) 



where 7 is the phase angle for the radical. 

This equation permits the calculation of the percentage of mechani- 
cal filtering developed at the film drum of the reproducer when a sine 
wave of velocity variation is applied at the sprocket. The calculated 



.80 


PT 1 






SINUSOIDftL VELOCITY ftPPLlEO W SPROCKET 




/f 






8 14 

| 

J> M 


/I 






1=ROTW?Y STftBILIZER 
2=EQUIVMENT DRMPED INERTm-ELhSTRKICE FITTER 
3'CQUIVrvLENT RESISTANCE -ELftSTRNCE FITTER 








4ljt 










V-4A 






YX. 


^ 


















\ 




. 


-~^^ 




















V 




^* 


-.. 


: -^. 








_ 


_ 




X. 

















FREQUENCY OF "WOW* 

FIG. 11. Transmission characteristic of film-moving system. 

performance of the system under these conditions is shown in curve 
1 of Fig. 11. The constants used in these calculations were those 
measured as described above. The characteristics of the equivalent 
damped inertia-elastance filter are shown in curve 2, where it is as- 
sumed that a flywheel equal to the stabilizer shell alone is fixed on 
the drum shaft and the damping resistance is made equal to that of 
the stabilizer. In practice, such damping would have to be added 
in some external manner that would not increase the steady film 
tension. This curve shows approximately the same resonance peak 
and practically coincides with curve 1 above this frequency, from 
which it may be deduced that the inertia of the inner flywheel does 
not contribute materially to the effective inertia of the shell, and 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 309 

hence does not enter into the filtering action except to provide a 
surface against which the shell may react. The results attainable 
from an equivalent resistance-elastance filter is shown in curve 3, 
in which it was assumed that no inertia appears upon the drum shaft 
and the damping resistance and film elastance are made equal to 
the constants used in the rotary stabilizer case. As in curve 2, it 
is obvious that this resistance would have to be obtained by some 
external means; and because of the assumption of constant elas- 
tance, the film tension can not be increased over that used in the case 
of the stabilizer. The filtering action is seen to be inferior to that of 
the stabilizer except in the region below approximately 3 cycles. 
Both filters would be so poor in this region as to be practically useless. 
The stabilizer has been designed so that this falls below the important 
range of frequencies for normal operation. 

It is of interest to determine the necessary ratio of inertia of the 
shell to that of the flywheel to permit the effective use of enough 
viscous resistance to establish the condition of critical damping. 
This may be done by equating to zero the discriminate of the re- 
duced cubic expression derived from the following equation. If the 
discriminate were made positive, the device would be over-damped. 
(pHLC + p*(l + L)Cr + pL + r] = (14) 

The reduction of this cubic is accomplished in the usual manner by 
assuming : 



and the result is: 

[x 3 + ux + v] = (16) 

where the constants u and v are functions of the circuit elements. 

For critical damping, the condition expressed by the following 
relation should exist: 

(l + 2^) = (17) 

If, in addition, it is assumed that t\ = (L/l), this becomes: 

[4C 2 (7? + l) 3 r 4 - r,HC{n* + 20r/ - 8\r* 4- 47? 4 / 2 ] = (18) 

or: 



h 2 4-207? - 8} =* Vrj(-n - 8) 3 \ (19) 

C V 8(77 + I) 3 / 

This result* states that the ratio, r;, of the inertia of the shell, 
/, and the flywheel, L, must be approximately equal to 8 or more, 

* This result was given by Hanna without proof in an unpublished memoran- 
dum. 



310 E. D. COOK [j. S. M. p. E. 

before it is possible to use sufficient viscous resistance to obtain 
critical damping. 

The transient solution for small shocks or impulses of velocity 
that are not great enough sensibly to vary the film loop elastance, is 
quite interesting. Naturally, such conditions are not maintained 
because the impulse is superposed upon the average or steady sprocket 
velocity, and this presupposes the existence of an oppositely 
directed impulse at some later time to return the system to its 
normal state. Since this merely involves the proper superposition 
of two similar but oppositely directed solutions, it is sufficient to 
investigate only the first impulse. This is a consequence of the 
assumed linearity of the fundamental equations. It must be realized 
that the solution will not apply to the case in which the film loop 
is disturbed by displacing it by the finger. Here the displacement 
is ordinarily made large enough to observe, in which case the elas- 
tance of the loops varies so greatly as to violate the conditions of the 
fundamental equations, which assumed constant coefficients. 

The resultant solution will be shown for the particular reproducer 
that was used in some of the test work and for which the measured 
constants have already been given. Since the discriminate of the 
fundamental cubic given below is negative, it is known that two of 
the roots are complex and the third is real. Practically, this means 
that critical damping is not attained. The correctness of the state- 
ment that this condition is not necessary in practice will be shown by 
the rapidity with which oscillations decay. 

[/> 3 + 8.33 2 + IQOp + 234 .4] = (20) 

The roots of this equation can be shown to be approximately : 

p, = (-1.5692) ) 

p 2 = (-3.3804 +j 11.741) > (21) 

ps = (-3.3804 - j 11.741) j 

Therefore, the transient part of the numerical solution for an 
impressed velocity of the unit function type (i. e., such that before 
the application of the shock, the velocity was constant and equal to 
the average velocity, i b , whereas after the instant application of the 
shock, the impressed velocity was constant and equal to i b -f- i), 
applied at the pulling sprocket, may be shown to be: 

i d = ai[l + 0.0739C- 1 - 5692 ' + 1.1151e- 3 - 3804 <cos(11.741* + 2.868)] (22) 

The transient response of the rotary stabilizer for unit impulses 
is shown in Fig. 12. The rapidity with which the oscillations decay 



Oct., 1935] 



THE HIGH-FIDELITY REPRODUCER 



311 



is well illustrated in this graph. The ordinates indicate the velocity 
of the film drum above its average velocity. The fact that the curve 
decays to a value of unity is the direct result of omitting that part of 
the impulse that would have existed in practice, and which would 
return the pulling sprocket to its average rotational velocity. The 
effect of a disturbance consisting of an instantaneous change of 
speed of the pulling sprocket, followed by a similar return to normal 
speed after any assumed time interval, /, may be readily determined 
by adding an identical curve of opposite sign to that of Fig. 12 but 
displaced along the time axis by the interval /. 




.& 7 .8 .3 

TIME (SECONDS') 



i.z 1.5 1.4 1.5 



FIG. 12. Transient response for unit function impulse. 

The advantage of this solution and its graphical representation 
is that it shows whether or not the particular design chosen is 
adequately damped. Likewise, it reveals the period of the natural 
oscillation of the system and, when combined with the data obtain- 
able from the steady-state filtering curve for sine wave disturbances, 
gives a rather complete insight into the performance of the device. 

It might be mentioned that these results have been checked by 
observations upon the natural period of oscillation and the rate of 
decay, and have been found to agree with the observations for the 
conditions under which they were calculated. Comparisons be- 



312 E. D. COOK [J. S. M. P. E. 

tween this reproducer and many others of different designs have 
been made over a period of several years of commercial operation, 
and in no case, so far, has equal performance in the matter of faith- 
fulness of reproduction or freedom from objectionable speed varia- 
tion been found. 

REFERENCES 

1 SHOWER, E. G., AND BIDDULPH, R.: "Differential Pitch Sensitivity of the 
Ear," /. Acoust. Soc. Amer., 3 (Oct., 1931), No. 2, p. 275. 

2 LAUTENSCHLAGER : Elektrische Nachrichten Technik, 11 (Dec., 1934), No. 
12, p. 409. 

3 CARSON, J. R.: "Notes on the Theory of Modulation," Proc. I. R. E., 10 
(Feb., 1922), No. 1, p. 57. 

4 VAN DER POL, B. : "Frequency Modulation," Proc. I. R. E., 18 (July, 1930), 
No. 7, p. 1194. 

6 RODER, H.: "Amplitude, Phase, and Frequency Modulation," Proc. I. R. E., 
19 (Dec., 1931), No. 12, p. 2145. 

6 WATSON: "Theory of Bessel Function," Cambridge Univ. Press, 2.22, p. 22. 

7 KELLOGG, E. W., AND BELAR, H. : "Analysis of Distortion Resulting from 
Sprocket Hole Modulation." Presented at the Spring, 1935, Meeting of the 
Society of Motion Picture Engineers, at Hollywood, Calif.; to be published in 
a forthcoming issue of the JOURNAL. 

8 LOOMIS, F. J., AND REYNOLDS, E. W. : "Development and Design of the 
High-Fidelity Reproducer." Presented at the Spring, 1935, Meeting of the 
Society of Motion Picture Engineers, at Hollywood, Calif.; to be published in 
a forthcoming issue of the JOURNAL. 

9 "Hydrodynamics," Research Council, Nat. Acad. of Sciences, 1932, p. 124. 

10 BERG, E. J.: "Heaviside's Operational Calculus," McGraw-Hill Pub. Co. 
(New York), p. 140. 

DISCUSSION 

MR. LUBCKE: Do I understand that the flywheel floats entirely free? 

MR. SCHULTZ: Yes, the inner flywheel floats as freely as possible, except for 
the coupling through the viscous oil medium. It acts merely as a surface against 
which reaction of the viscous medium can take place, and as long as its inertia 
is sufficiently large that it moves essentially at constant speed the design consid- 
erations are fulfilled. 

MR. FRAYNE: Does the passage of splices affect the motion? 

MR. SCHULTZ: The passage of a splice would correspond to a small transient 
disturbance, which would be very rapidly damped; the amplitude of operation, 
due to the design constants, is sufficiently small that it is not perceived by the ear. 

MR. TASKER: Referring to Fig. 10, I suspect that, in common with many 
others, it was Mr. Lubcke's first impression of this circuit that the effectiveness 
of the filter must be very much less than it would be if the large flywheel were 
solidly coupled to the shaft and thus used in the orthodox manner. That such is 
not the case may perhaps be better understood from the following brief, though 
non-rigorous, comments. 

In Fig. 10 we are concerned with the current id, which corresponds with the mo- 
tion of the film over the drum, to which drum is firmly attached the mass of the 



Oct., 1935] THE HIGH-FIDELITY REPRODUCER 313 

outside shell of the rotary stabilizer. The electrical equivalent of this shell is 
properly shown as the inductance I. It is quite true that the reactance of this 
element is small compared to that of the large floating mass represented by L. 
It is also true that speed variations impressed upon the filter encounter, in addi- 
tion to the shunt reactance C, only the small reactance of / and the comparable 
resistance r, and are not affected by the much greater positive reactance of L. 

In spite of the fact that a comparatively large amount of "flywheel effect" 
residing in L is not effectively used, the filter may still be effective if the value of C 
is made large enough. It is obvious from the diagram that the amount of alternat- 
ing (as speed variation) component appearing in id depends upon the relative 
impedances of the shunt branch C and the series branch / -f- r at any ''wow" 
frequency for which the reactance of L is large, and hence provision of a very large 
C(e. g., very soft film loop) may result in good filtering. Now, if the resistance r 
were omitted, so that the very large positive reactance L -f- / became effective 
in the series branch, the filter would be undamped, and to attain corresponding 
damping, the appropriate resistance would be inserted either in series with the 
positive reactance L + / or in series with the negative reactance C. In either event 
critical damping will require that this resistance be proportional to the square 
root of the ratio of the effective inductance divided by the effective capacity, and 
hence must be much larger than the small resistance r shown in Fig. 10. If placed 
in series with the inductance, the steady component of id must flow through this 
resistance, resulting in greatly increased film tension (hence smaller C} ; and if 
placed in series with capacity C, it will increase the impedance of this shunt 
branch to such an extent as to offset the increased impedance of the series branch 
L -f- /. In consequence, the benefits of the much greater series element, L, is by 
no means fully realized, or may even result in a penalty. 

A further study of Dr. Cook's very thorough analysis will point to more precise 
conclusions. It has been my purpose only to indicate the general effect of this 
new design as compared to a more conventional type. 

MR. KELLOGG: The ball bearing that carries the flywheel inside the shell 
does not run continuously. Once the machine is up to speed there is no relative 
motion between the flywheel and the shell except for what speed changes the shell 
undergoes. There is no bearing that does not have some initial breakdown torque 
with even a light load. 

When this machine was proposed I was skeptical as to whether we could make 
the friction of the bearing upon which the free-running flywheel runs, low enough 
to insure that the flywheel and the outer shell would not practically lock together 
when the amplitude of oscillation fell below a certain critical value, thereby pre- 
venting any relative motion and likewise preventing damping. That such a condi- 
tion is not encountered has been established by numerous tests. If the flywheel 
and shell are locked together the action is very different, and a small disturbance 
starts an oscillation that persists for a long time. 

We have not been able to observe any oscillation of such small amplitude that 
relative motion and consequent damping do not occur. It seems to me that the 
reason may be that the radial force or load upon the bearing is continually changing 
direction relative to the bearing. A slight irregularity in the bearing which might, 
in the case of an ordinary bearing, prevent relative motion when a small force is 
applied, is relieved and the bearing freed when it is turned the other side up. 



NON-THEATRICAL PROJECTION 
R. F. MITCHELL** 



Summary. Various factors involved in projecting 16-mm. pictures for industrial 
and educational purposes are fully analyzed, according to three main divisions: 
(1) the illumination of the screen and how it is modified, (2) the effectiveness of pro- 
jection as regards the audience, and (3) miscellaneous related factors. Each broad 
division is broken down into sub-divisions enumerating the principal items involved. 
Curves show the relative cost of attaining desirable screen illumination with various 
lamps, and the relations between lumens, foot-candles, and screen sizes. 

Based upon the analysis, recommendations as to a standard method of measuring 
screen illumination, and a list of suggested screen sizes for various lamps and lenses, 
are given as having been found satisfactory for ordinary work. 

INTRODUCTION 

There has been a vast increase in the use of motion pictures in recent 
years for industrial, educational, and similar purposes. The most 
outstanding progress has been in the application of 16-mm. equipment 
in industry and education so much so that 16-mm. is no longer a 
purely amateur standard but fills a definite sphere of usefulness all its 
own. In fact, 16-mm. projection has reached the stage where the 
term "sub-standard" conveys a totally different meaning from the 
rather "off "-standard implication it used to convey. The advance 
has been due to the high technical and mechanical excellence of the 
equipment available ; improvements have been made so rapidly that 
it is hard to keep up with them. It is not surprising, therefore, that 
so much interest has been evinced in trying to tabulate the material 
available, in formulating consistent methods of judging the possi- 
bilities, and in evaluating the problems of the field. In its broader 
aspects, this is the function of the Non-Theatrical Equipment Com- 
mittee, and involves photography, projection, printing, etc., as with the 
35-mm. film. It involves, also, reduction printing, reversal and other 
technical aspects, in addition to problems arising from the fact that 
non-theatrical equipment is used by amateurs. This paper is an 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Bell & Howell Co., Chicago, 111. 

314 



NON-THEATRICAL PROJECTION 315 

attempt to enumerate and analyze the various factors involved in 
non-theatrical projection. Although mainly concerned with 16-mm. 
problems, the essential data and conclusions are valid for film of any 
size and are applicable in many respects even to purely theatrical 
projection. 

Analysis of the Non-Theatrical Problem. The sharpest dividing line 
between theatrical and non-theatrical projection is in respect to the 
size of the film used and the fact that incandescent lamp illumination 
is used almost exclusively for non-theatrical work. This tends to 
limit somewhat the maximum size of satisfactory picture attainable, 
but the difference is nowhere as great as it is generally regarded. 
This is due to the powerful lamps now available with 16-mm. ma- 
chines and to the efficient mechanisms and fast lenses now character- 
istic of such units. The discussion necessarily covers the more power- 
ful and more efficient models of 16-mm. projectors such as would be 
used for the industrial and educational work. The requirements of 
these fields have been in mind in working out this paper. It is 
convenient to divide the analysis into two main divisions: (7) the 
projection problem (the picture on the screen) ; and (2) the audience 
problem (the efficiency of the projection as far as the audience is 
concerned). Each of these main divisions may be broken down as 
follows : 

(1) The Projection Problem 

(a) Efficiency of the intermittent movement. 
(6) The optical system. 

(c) The lamp. 

(d) The picture size. 
(2} The Audience Problem 

(a) Screen reflection characteristics. 

(b) Screen deterioration. 

(c) Auditorium illumination level. 

(d) Film density. 

(e) Color. 

(,?) Miscellaneous Related Factors 

(a) Flicker. 

(b) Steadiness. 

(c) Visual contrast and resolution of detail. 

(1) THE PROJECTION PROBLEM 

(a) Efficiency of the Intermittent. The intermittent mechanisms 
employed in 16-mm. projectors are, in general, more efficient than 



316 R. F. MITCHELL [J. S. M. P. E. 

those employed in 35-mm. equipment. The shorter stroke and lower 
inertia permit a faster acceleration and deceleration. In addition, 
there is more than one intermediate flick for each frame, in most 
16-mm. projectors, so flicker is reduced (this is covered further in sec- 
tion 3a). Even so, the average efficiency of 16-mm. projector inter- 
mittents can be regarded as being approximately 60 per cent, as 
compared with about 40 per cent for the average theatrical projector. 
Although the efficiencies of the diiferent makes of 16-mm. projectors 
vary, they are all close enough to 60 per cent that the differences need 
not be considered in this analysis. It is sufficient to make the point 
of the higher efficiency of 16-mm. projectors in general. 

(b) The Optical System. No 35-mm. projectors are optically as 
efficient as the 16-mm. projectors. Thirty-five-mm. projectors, 
especially portable projectors, employ //4.5 to //2.3 lenses, whereas 
most 16-mm. projectors use lenses varying from //2.5 to //1. 6 

TABLE I 

Relative Light Transmission 



Focal Length 
//Number 


0.64 3/4 
4 2.5 


1 

2.6 


1V2 

1.8 


2 
1.65 


2 
2.1 


21/2 

1.9 


3 
2.3 


3i/ 2 

2.7 


Relative 
Light 
Trans- 
mission 


1 2.5 
1 


2.4 
0.9 
1 


5 
1.9 
2.1 
1 


6 
2.3 
2.5 
1.2 


3.6 
1.4 
1.5 


4.4 

1.7 
1.9 


3 

1.2 
1.3 


2.2 

0.8 
0.9 



on the average, three times as fast as the equivalent 35-mm. portable 
and at least P/2 times as fast as the best theatrical projection equip- 
ment. This additional optical speed, in conjunction with the high 
intermittent efficiency, explains why 16-mm. projectors can project 
such large pictures in comparison with 35-mm. portable projectors 
of equivalent lamp wattage. The optics of 16-mm. projectors vary 
within considerably wide limits, but as far as the projectors that 
would be used for serious work are concerned, the fastest optical 
combination is about 2 times, perhaps 3 times, as fast as the slowest 
16-mm. industrial projector combination enough to warrant further 
analysis. 

For most industrial purposes the regular 2-inch lens can be used, 
having an average speed of about //1. 65. However, for auditorium 
projection involving long throws, longer focal length lenses are used-^ 
3, 3 Ye, or 4 inches. The average speed is about //2. 7 which is 2 x /2 
times as slow as the 2-inch, //1. 65 lens. On the other hand, for con- 
tinuous projection with display cabinets, smaller screen sizes and, 



Oct., 1935] 



NON-THEATRICAL PROJECTION 



317 



usually, shorter focal length lenses are employed, because of the neces- 
sity of counteracting daylight or other light incident upon the screen. 
These relations are covered by Tables I and II. 

TABLE II 

Width of Picture Attained with Filmo Projection Lenses 



Distance 
from 
Screen 
(Feet) 


Focal Length of Lens 
(Inches) 


0.64 


3 /4 


l 


11/2 


2 


2V* 


3 


31/2 


4 




ft. in. 


ft. in. 


ft. in. 


ft. in. 


ft. in. 


ft. in. 


ft. in. 


ft. in. 


ft. in. 


1 


7 


6 
















2 


1 2 


1 


9 














3 


1 9 


1 6 


1 1 


9 












4 


2 4 


2 


1 6 


1 


9 










5 


211 


2 6 


1 10 


1 3 


11 










6 


3 6 


3 


2 3 


1 6 


1 1 


10 








8 


4 8 


4 


3 


2 


1 6 


1 2 








10 


510 


5 


3 9 


2 6 


1 10 


1 6 


1 3 


1 




12 


7 


6 


4 6 


3 


2 3 


1 9 


1 6 


1 3 


1 1 


16 


9 4 


8 


6 


4 


3 


2 4 


2 


1 8 


1 6 


20 


11 8 


10 


7 6 


5 


3 9 


3 


2 6 


2 1 


1 10 


25 


14 7 


12 6 


9 4 


6 3 


4 8 


3 9 


3 1 


2 8 


2 4 


32 






11 11 


8 


6 


4 9 


4 


3 5 


3 


36 






13 5 


9 


6 9 


5 4 


4 6 


3 10 


3 3 


40 






14 11 


10 


7 5 


6 


5 


4 3 


3 9 


45 








11 3 


8 5 


6 9 


5 7 


4 9 


4 2 


50 








12 6 


9 4 


7 6 


6 3 


5 4 


4 8 


64 










11 11 


9 7 


8 


6 11 


6 


75 










14 


11 3 


9 4 


8 


7 


100 










18 9 


15 


12 6 


10 8 


9 4 


125 










23 5 


19 8 


15 7 


13 4 


11 8 


150 










28 1 


22 5 


18 8 


16 


14 



(c) The Projection Lamp. For many kinds of industrial work, 
especially continuous, the optics and screen size are fixed, so that any 
desired increase of illumination must be attained by some other 
means. This problem came up in practically all such continuous 
installations at the recent Century of Progress Fair at Chicago, so 
the discussion of this aspect will include other conditions in which 
additional illumination is attained by using more powerful projec- 
tion lamps or by modifying the existing lamp set-up. Obviously, 
the use of a more powerful lamp is the first alternative; but beyond 
that, further increase of illumination can be attained within certain 



318 



R. F. MITCHELL 



[J. S. M. p. E. 



limits by over-voltaging the lamp. In such cases the additional cost 
due to shortening the lamp life is regarded worth while. Conversely, 
when a powerful projector is used for showing pictures upon com- 
paratively small screens, economies can be effected by operating the 
lamp at voltages below normal. These relations are shown very 
clearly in Fig. 1. These curves must be used in combination with 
the table accompanying them, because the price of lamps bears no 
comparative relation to the emitted light or the life. The curves 
warrant careful study, however, because they show how much more 
economical it is to use a high-wattage projector at low voltage. For 



*r 


HOU $ 

9?^ 
S 3 

44.5 


VOLTS 

fexcfu-r 

92 
93 
94 
95 

9i 


-v 


* 

I 


ro 






t 






UK C 


OST 















/ 000 IV. 

71. S 
74 

60 
S3 


750 W. 

So.S 
s^.2 

S4A 
S6.S 
S3. 7 


SOO*. 

32 4 

33. S 

36.Z 
37.6 
39.Z 


&<t.S 
93 
96 S 

\ioo\ 

/03.S 
107 5 
III.S 


3.3 
65.7 
66.2 

73.2 
76 
766 
61.3 
84.1 


40.6 

46.6 

40.7 
SO.S 
SZ.I 
S3 9 


J 
? e 

z 

1 3 

1 
1 & 

1 3 


<?& 

IQQ 
101 

103 

IOS 






\ 


\ 


\ 

S 


N^ 

S 

afo 1 


^r 

^ 


g 


^ 
















17?. S 


66 6 


SS 5 
















S 


^ 














126 S 


ft 5 




/ 3 
















"V 


~Xo 


"" 


*^^^ 














g 


S9.4- 




f O Q 
















s^ 


f. 


~\^ 




"~-~^ 











135 
/3<f ] 


9f.S 

9 V 


61. Z 
63 




/Of 

no 














^0 


^^- 


^^= 




^^ 


~-^ 





FIG. 1. Lamp-hour cost chart : 1000-w. lamp used as basis of comparison, 
and assuming same screen size and projector used with the three sizes of 
lamps shown in table. 

instance, a 1000- watt projector at 90 per cent voltage provides 68 
per cent of the screen light at a cost of 10 cents per hour. (For con- 
venience, the 1000-watt projector at normal voltage is assumed as 
providing 100 per cent of screen brilliancy.) 

To obtain the same intensity of light with a 750-watt projector, 
the lamp would have to be operated at 99 per cent of the normal 
voltage at a cost of 22 cents per hour twice as much. Even operat- 
ing the 750-watt projector at 10 per cent overload, the screen illumi- 
nation Would still be slightly lower than the 1000-watt projector 
operated at normal, and the cost is $1. 13 per hour as compared to $0.43 
per hour. However, if only a 750-watt projector is available, it is of 
great help to know that the equivalent illumination of a 1000-watt 
machine can be attained. If the projector is overloaded for periods 



Oct., 1935] NON-THEATRICAL PROJECTION 319 

shorter than the full lamp life, the comparative cost would be less, as 
the figures are necessarily based upon the assumption of operating the 
lamp for its full life at the voltage indicated. It is quite evident that 
the screen illumination from any given projector can be varied within 
quite wide limits according to what lens and lamp combination is 
employed practically all 16-mm. industrial projectors are designed 
so that lenses and lamps can be interchanged. Therefore, it seems 
logical to outline a standard set of conditions under which the 
normal screen lumen rating of any projector may be determined. 
With that established, we are in a much better position to apply 
the special modifications covered above. The following procedure 
was used to establish the data given in Fig. 1 and is offered as a con- 
tribution toward a Society recommendation for Standard Practice in 
this field. 

PROPOSED STANDARD METHOD OF DETERMINING THE SCREEN LUMEN 
RATING FOR 16-MM. PROJECTION 

The parts involved in the process of illuminating the screen are 
the lamp, condenser, shutter, and lens; and their efficiency, sepa- 
rately and in combination, determines the relation between the amount 
of light reaching the screen and the total spherical luminous energy 
emitted by the lamp. The screen lumen output value of a pro- 
jector is a measure of its projection illumination capacity. This 
value remains the same (within reasonable limits) regardless of the 
screen size. The following procedure is suggested for determining 
the screen lumen rating of a projector : 

(1) Make the Test with the Projector Running (at Approximately 
Normal Speed) without Film. 

(2) See that All Optical Parts Are Scrupulously Clean. 

(3) Use the Lens and Condenser Regarded as Standard Equipment. 
When other than standard equipment (2-inch focal length) lenses 

are used, the projector may be found to produce more or less illumina- 
tion 'than with standard equipment. The special lens may be of 
larger or smaller aperture, or the condenser may be more or less 
efficient. 

(4) Use a Rated Lamp, at the Correct Voltage. 

A rated lamp is one that has been checked by its manufacturer 
as to its efficiency of operation at various voltages in relation to its 
normal efficiency at the design voltage and as to its design life. A 



320 R. F. MITCHELL [J. S. M. p. E. 

rated lamp can be operated at the design voltage in a brief test with- 
out serious loss of accuracy; but as the test period is usually extended, 
and usually more tests than one are required of a lamp, it is better 
to use a lamp rated at a voltage that is about 80 per cent of the design 
voltage. The foot-candle reading is then multiplied by the conver- 
sion factor for the particular lamp. The exact voltage and the con- 
version factor are supplied by the lamp manufacturers, and are 
marked upon the lamp. 

(5) Use a Screen Size That Yields an Intensity of About 6 to 10 
Foot-Candles. 

This tends to eliminate errors in reading the foot-candle meter. 

(6) Take Five "Foot-Candle Readings," One at the Center, One Each 
at the Centers of the Side Edges, and One Each at the Centers of the 
Top and Bottom Edges. 

The average of five foot-candle readings taken at the points 
mentioned will be so nearly the exact average foot-candle value for 
the screen that a larger number of readings will not be necessary. 

In 1920, W. F. Little published 1 two sets of screen foot-candle 
readings: one set giving the light intensities at sixteen points of the 
screen and the other giving readings at 256 points. The latter table 
is reproduced as Table III. It is significant that the average of the 
sixteen readings in Little's first table is 10.0 foot-candles and the 
average of the 256 readings is 10.1 foot-candles. In Table III 
nine readings have been set in bold-faced type that seem to represent 
fair intensities at those points. The average of the readings at the 
five points recommended is 10.0 foot-candles, while the average of 
the nine points is 8.6. Experience indicates that the nine-point 
reading usually made in theatrical work gives an average lower than 
that actually realized. It is obvious that any discussion of screen 
illumination (which is obtained by multiplying the average foot- 
candle intensity by the area in square-feet) must recognize the neces- 
sity of standardizing the manner of determining the average in- 
tensity. 

It is suggested that the subject be considered further by the 
Projection Screen Brightness Committee and the Standards Com- 
mittee. Also, it is recommended that the illumination at the center 
of the picture be not more than 15 per cent greater than at the top 
and sides. 2 



Oct., 1935] NON-THEATRICAL PROJECTION 



321 



rH c o> r- cc 

COOOOiOO'-iOai'-iOOCOi-HC'OCOT-iCiOO 

dddd'-idd'-Hdd^da>a>r--d 

TH_|,-|,-H,-li-lT-lTH,-H,-|,-lT-H 

00 (M <M LO LO O T-H Tti Tfri LO i i O CO ?O Tfi i i o 

O^T-irHi-Hr-(T-H^^-i^-!rH^doio6l> '"I 
^H ,-H ^-1 t-H T~^ iH i~~i ^^ r f T-H r ( T I T I ^^ 

T 

COcOOO'-iOOiI>I>cOOO>OT)Hl>(Mi ib- OJ 
........ two 

C^r-i^-lr-l^-I^H^-lrHOOOit^ G 

i-H H il T t i I r-H T I T H T-H f-H rH i-H T-H r-H *^^ 

_ T3 

S 

^> moOCO^OlOOO^Oi' iLQCOt-^iOcO'-* *" 

1 ^ ............... co 

e i-^^nr-ic^ I<MC<IHT icqi IT loooioo o 

^ ^H^^^^rH^^^H^rH^rH^H CK| 

42 "o 

8 Ttit^r^rticoocoi>c>.T icoor^-rjHTtifM <u 

g ^HT-!I l(MC^(N'-Hi IrHCqi IrHOdoiOO ^ 

1 ' s 

<JH 

T-lT-(-COiCOCO'*COCOCO' (Olt^CO'-HOi 
!^ ^H^-(T-HT-(T-(T-Hi-(rHrHT iT-HOOOOSr^. 

O5'-i<MCO^>O'*iiOTt<C<lOCOCOO5COO5 
Oi ii ii i i-( T I^-I^HT-HT-II iOOO5Oi!> 

I>O' i iO <M ^ O 1-1 CC i COCO' (COOO^f 

O rH r-l rH l-l T-H i I i lr-lr-(OOOO5G6l> 



oooooooocic 



322 R. F. MITCHELL [J. S. M. p. E. 

(7) Multiply the Average of the Five Readings by the Lamp Conversion 
Factor, if Necessary. 

If the lamp is undervoltaged the reading must be multiplied by the 
conversion factor in order to obtain the value of the design voltage. 
As there is considerable variation in color distribution between the 
outputs at the two voltages mentioned, as well as some variation 
among different types of lamps at the design voltage, a meter filtered 
and graduated to correspond to the sensitivity curve of the human 
eye is most desirable. (A meter not thus equipped will require the 
use of further individual correction factors for obtaining accurate 
values.) 

(8) Deduct the Room Illumination in the Region of the Screen. 

If a black-walled room is available for the test, no deduction is 
necessary and a more accurate test can be expected. 

(9) Reduce This Remaining Value by a Factor Representing the 
Average Efficiency of the Lamp for Its Life Period. 

As the lamp output continuously decreases during use, the average 
output during the lamp's lifetime becomes the best measure of the 
illuminating value. This falling-off is due to the gradual blackening 
of the bulb and the gradual increase of electrical resistance as the 
tungsten evaporates from the surface of the filament. For the usual 
25-hr, (approx. 3300 K) lamp, the average lifetime output of prevailing 
types of lamps is, for the biplane filament, about 86 per cent of the 
new lamp output, and about 93 per cent in the case of a monoplane 
filament. These figures are approximate ; the percentage varies with 
the type of lamps. 

(10) Multiply This Reduced Value by the Area of the Screen (in Sq. 
Ft.) to Obtain the Screen Lumens Value. 

This value represents a true measure of the average screen-illumi- 
nating capacity for the life of the lamp. When it is compared with 
the total spherical lumens output of the lamp (also lifetime average), 
the efficiency of projection is indicated. 

The output of the rated lamp matches only the average output of 
a large number of stock lamps; therefore, it is evident that a stock 
projector with a stock lamp is not likely to match the rated lamp as 
to screen lumens. The lamp manufacturing tolerance, in watts, 
for stock projector lamps is 5 per cent, the equivalent of which 



Oct., 1935] 



NON-THEATRICAL PROJECTION 



323 



in lumens is 12 per cent. Voltage variations of 5 per cent in 
projection may add another =*= 18 per cent in lumens. 

(d) The Picture Size. It is apparent that any given projector 
gives just so much total light. By placing the projector near the 
screen or far from it, or by using lenses of various focal lengths, the 
image brilliancy can be varied. The foot-candle intensity is a direct 
function of the screen size, and can, therefore, be shown in con- 
venient chart form as in Fig. 2. The lumen output of any projector 
being known, it is a simple matter to determine the size of the picture 



14 lj 1 1 Z 23. ZJ zl 3 Jf 3? 3 f*7j ?^ if ^. /. y.^ 




FIG. 2. Chart of screen illumination and screen width. 



for any desired intensity. Experience indicates that an intensity of 
6 foot-candles is quite adequate for non-theatrical black-and-white 
projection. The points of view of several projector manufacturers 
on this score will be found in the Report of the Committee on Non- 
Theatrical Equipment. 3 In this connection it is interesting to note 
that an intensity of 6.9 foot-candles afforded "very comfortable" 
projection, of "good" visual acuity, and that an intensity of only 3.8 
foot-candles was rated "comfortable" and "fair" in a report of the 
Theater Lighting Committee. 4 The Projection Screens Committee's 
preference 5 of 7 foot-lamberts illumination is also of interest even 
though it refers to arc illumination. 



324 



R. F. MITCHELL 



(2) THE AUDIENCE PROBLEM 



[J. S. M. P. E. 



All the material presented above covers the illumination reaching 
the screen. As a matter of convenience, such tests are made with the 
machine running without film, and the resulting illumination is 
measured at the screen. The Weston photoelectric foot-candle 
meter used in the tests is probably as convenient and as reproducible 
a standard as available. The photo-cell should be fitted with a filter 
for translating the readings to the visual response curve. This filter 
is especially valuable because it corrects for a wide variation of color 
temperature. 6 

(a) Screen Characteristics. A point that seems to be overlooked 



300 




O ZO 

OF OBSERVATION 



FIG. 3. Brightness characteristics of Da-Tone perforated motion picture 
screens: solid curve, type Z screen; dot-and-dash curve, beaded screen; 
dotted curve, silver screen. 



or, at least, not considered as fully as its importance seems to war- 
rant, is that concerning the effective image brilliancy as regards what 
the audience sees. Everyone realizes that the efficiency of the screen 
varies that is the main reason why the convenient method of 
measuring at the screen is used so widely. At the same time, the 
screen reflection factor should be checked, as well as its polar distri- 
bution (Fig. 3). The percentage reflection of the screen can easily 
be determined by merely taking a series of readings with the photo- 
cell turned toward the screen as well as with the photo-cell facing the 
projector. (Obviously the photo-cell must not throw a shadow upon 
the screen. Such a test is best made with the photo-cell about a foot 
or so away from the center of a fair-sized screen.) This will give the 
intensity of the light reflected at right angles to the screen; the dif- 



Oct., 1935] NON-THEATRICAL PROJECTION 325 

ference at various angles can most readily be calculated from curves 
supplied by the screen manufacturers, such as those in Fig. 3. For 
example, suppose on a standard size of screen an average reading of 
6 foot-candles was obtained with the photo-cell facing the projector 
and 4.5 foot-candles with the cell facing the screen. This would 
indicate a screen reflection efficiency of 75 per cent at right angles. 
Viewing a beaded screen at a 10-degree angle the audience would 
see an image of intensity 160/270 of 4.5, or 2.67 foot-candles. This 
probably represents about the poorest condition encountered in 
average non-theatrical projection of acceptable quality. It is 
recognized, however, that the term "acceptable quality" is rather 
loosely used ; this paper is written with a view of assisting to establish 
a workable definition or specification of this condition. 

(b) Screen Deterioration. As is well known, the screen reflection 
efficiency is not at all stable, but changes with the age and use of the 
screen. That is one of the main reasons why it has become customary 
to read initial "screen" brilliancy instead of "audience" brilliancy. 
It would therefore seem necessary in any adequate specification to 
indicate some acceptable average screen efficiency, measured at 
right angles to the incident beam. A factor of 75 per cent is suggested 
as a basis, keeping in mind the high efficiency and comparatively good 
cleaning qualities of beaded screens, which seem to be used most 
widely for non-theatrical work. Moreover, screens used for non- 
theatrical work are not perforated, for which reason they have a 
higher initial efficiency than theatrical perforated screens, amounting 
to about 6 or 8 per cent. 

(c) Auditorium Illumination Level. The efficiency of projection, 
so far as the audience is concerned, depends also upon how much or 
how little extraneous light there is in the hall to conflict with the 
light coming from the screen, as well as cigar and cigarette smoke 
which is occasionally a factor in non-theatrical work. Smoke 
diffuses the light throughout the room, in addition to cutting down the 
initial light reaching the screen. 

(d) Film Density. The film density affects the quantity of light 
reaching the audience. This factor is difficult to standardize for non- 
theatrical work as it seems to vary more than in theatrical release 
printing. Sixteen-mm. non-theatrical films can be divided into the 
three broad classes of direct reversal, 16-mm. contact, and 16-mm. 
reduction prints. 

(e) Color. Color should probably be added now as an additional 



326 R. F. MITCHELL [J. S. M. p. E. 

classification, because there commended screen brilliancy will prob- 
ably be different from what is required for black-and-white. Because 
color pictures have a color contrast usually considered in black-and- 
white work, a lower relative screen brilliancy is generally acceptable 
for color than for monochrome pictures. This, fortunately, offsets 
to some extent the loss of light due to the absorption of the colors in 
the film. Existing concepts of color absorption will have to be 
modified now that the 16-mm. Kodachrome process is available, 
but it should be possible to establish some recommendation for 
desirable screen intensity. Due to the high transparency of the 
colored image of the Kodachrome film, and remembering the gain 
in visual acuity due to color contrast, it would seem that the recom- 
mended intensity for black-and-white work would be adequate, 
although perhaps a somewhat higher intensity, say, 8 to 10 foot- 
candles, might be desirable. We have not had sufficient experience 
with the new color process to permit definite recommendations, but 
believe that the screen sizes suggested in Table IV will be found 
satisfactory. 

(3) MISCELLANEOUS RELATED FACTORS 

It is rather painfully evident that existing methods of rating or 
regarding projection efficiency have all been from the point of view of 
the machine, lamp, and screen efficiencies. The basic physiological 
factors are not adequately known, and so are considered only from 
a more or less empirical standpoint. Strictly speaking, we should 
start from the eye of the observer and work back to the screen, and 
establish our recommendations from that point of view. Some of 
the factors to be considered in this connection are as follows : 

(a) Flicker. Flicker is a function of the speed of the projector and 
of the extent and number of flicks during the projection of each frame. 
In general, with 16-mm. projectors, it is less important than in theat- 
rical work (see section la). Visual acuity is improved appreciably, 
and there is a lower fatigue factor, 2 ' 7 both matters of extreme impor- 
tance in educational work. 

(b) Steadiness. As far as we know, 16-mm. projection (assuming 
the higher grades of projectors are used) is steadier than theatrical. 
Travel-ghost, or improper shutter timing, is practically unknown. 
A recent War Department specification called for a maximum image 
jump of L /4 inch upon a screen six feet high. In really high-grade 
16-mm. projectors the jump upon a 6-foot screen is less than 1 / 8 



Oct., 1935] NON-THEATRICAL PROJECTION 327 

inch. 2 This is also most important in educational work on account 
of the lower fatigue involved in viewing rock-steady pictures. 

(c) Visual Contrast and Resolution of Detail. The effective con- 
trast of the picture as seen by the eye depends upon all the factors 
involved in projection, and it is accordingly very difficult to define 
or specify what acceptable visual contrast may be. It depends, 
first of all, upon the color as well as upon the intensity of the light. It 
depends upon whether the film is colored or whether the silver image 
in a black-and-white film has been stained. It depends upon the 
grain structure and the resolution of detail in the print, which, in 
turn, are affected by the contrast characteristic of the lens. It 
depends also upon the color of the screen and the detail resolving 
power of the screen. The last factor is further dependent upon the 
distance of the observer as related to the magnification of the image. 

CONCLUSION 

Due to the multiplicity of the factors involved in any compre- 
hensive discussion of the visual efficiency of projection, the accepted 
method of measuring screen illumination is regarded the most con- 
veniently satisfactory method for general use provided that the 
measurement is made under clearly defined conditions and that the 
screen characteristic be considered more fully than usual, at least in 
non- theatrical work. The recommendations in Table IV for suit- 
ably sized beaded screens for ordinary purposes are the results of 
extensive consideration of the many factors outlined above. 

TABLE IV 

Recommended Screen Sizes 

Wattage Size of Screen 

of Projector Lens Size of Screen Recommended for Kodachrome 

Lamp Used Speed Recommended for B & W (Tentative) 

(//Number) 

1000 1 . 65 6X8 feet 5X7 feet 

750 1 . 65 5X7 feet 4 1 /* X 6 feet 

750 2 4 J /2 X 6 feet 3X4 feet 

500 2 3X4 feet 30 X 40 inches 

400 2 30 X 40 inches 22 X 30 inches 

300 2 22 X 30 inches 18 X 24 inches 

These screen sizes are suggested upon the basis of a screen illumina- 
tion of the order of 6 foot-candles, with the projector running without 
film and fulfilling the ten requirements as detailed above. The sizes 
given in the table can be doubled for special cases. It is recom- 



328 R. F. MITCHELL [J. S. M. P. E. 

mended also that projectors suitable for serious non-theatrical work 
project steady pictures with less than x /4 inch of jump in a picture 
six feet high. Also, it is recommended that the illumination be 
uniform over the entire screen, within 15 per cent. 

REFERENCES 

1 LITTLE, W. F.: "Tests of Screen Illumination from Motion Picture Pro- 
jectors," Trans. Soc. Mot. Pict. Eng., IV (1920), No. 10, p. 38. 

2 PANDER, H.: "The School Projector, How It Should Be," Film fur Alle 
(March 3, 1935), p. 81. 

* Report of the Committee on Non-Theatrical Projection. Presented at the 
Spring, 1935, Meeting; to be published in a forthcoming issue of the JOURNAL. 

4 Report of the Theater Lighting Committee, /. Soc. Mot. Pict. Eng., XVI 
(Feb., 1931), No. 2, p. 241. 

6 Report of the Projection Screens Committee, /. Soc. Mot. Pict. Eng., XVIII 
(Feb., 1932), No. 2, p. 247. 

6 GOODWIN, W. N., JR.: "The Photronic Illumination Meter," Tran. Ilium. 
Eng. Soc., 27 (Dec., 1932), No. 9, p. 828. 

7 SNELL, P. A.: "An Introduction to the. Experimental Study of Visual 
Fatigue," /. Soc. Mot. Pict. Eng., XX (May, 1933), No. 5, p. 367. 

DISCUSSION 

MR. CRABTREE: It would seem from these data that it is highly desirable to 
designate the potential illumination by screen lumens rather than by wattage. 
In other words, if you have a 500-watt lamp with a very inefficient optical system, 
it might not give you as high a screen brightness as, perhaps, a 250-watt lamp 
with an efficient projector. 

MR. DUBRAY: That is true, and the evident scope of the paper is to suggest 
such a procedure. 

MR. FARNHAM: The matter of operating a higher wattage lamp at a lower 
voltage in order to gain lamp life, and still apparently have enough screen il- 
lumination has rather interesting possibilities. It seems to me that if you can get 
along with the fullest output of the lamp, it might be better to reduce the cost of 
the projector; in other words, to buy a smaller projector with a smaller lamp, 
and operate the lamp at full brilliance. My experience shows that it is generally 
better to control the light at the projector, and operate it at full brilliancy. 
Some screen illumination tests made a few years ago showed that when the 
observer adjusted the amount of light, in almost every case he took all the light 
he could get. 

The table in your paper must include lenses along with the lamp wattage, be- 
cause the lens has considerable effect upon the screen illumination. Projectors 
vary as greatly as two to one, as to light from a given lamp, so I don't believe an 
evaluation by wattage of the lamp will work out as well as one in terms of the 
lumen output of the projector. 

MR. DUBRAY: No true evaluation of light efficiency can be made if the whole 
optical system of the projector is not taken into consideration. 



Oct., 1935] NON-THEATRICAL PROJECTION 329 

As to regulating the screen intensity by undervolting the lamps, although Mr. 
Farnham's remarks are very true, it must be borne in mind that the portability 
of 16-mm. projectors prompts their use under greatly varying conditions. In 
industrial or classroom or school auditorium projection, one may have to project 
one day in a hall where the projector is set 100 or more feet from the screen; 
the next day or in another classroom, the same operator may have to use the same 
projector for 20- or 30-foot throws. Control of screen illumination through volt- 
age control is most desirable under such conditions. 

MR. FARNHAM: Under such conditions the machine might be operated by a 
skilled operator who knew what was involved; but as a general recommendation 
to all users of projectors, I am afraid more difficulties may be caused than good 
accomplished. 

MR. DUBRAY: We have found, however, a great response on the part of users 
of our equipment in assimilating the rather simple fundamental principles that 
govern good projection. 

MR. SHAPIRO: The cost per hour of the 1000-watt lamp, giving il- 
lumination equal to that of the 750-watt lamp, indicated that it would be more 
economical to use the higher-wattage lamp underloaded in order to effect lower 
lamp hour cost. That is particularly interesting in connection with the general 
complaint about the high original cost of these projection lamps, and may be a 
deciding factor in projector design, particularly where the requirements are 
flexible, such as in the case cited by Mr. Dubray, in which the projector was used 
with a throw of 100 feet and then with one of 25 feet. 

MR. FARNHAM: In Fig. 1 the light of the 500-watt lamp was shown to be 
only about 45 per cent of that of the 1000-watt lamp. In other words, the in- 
crease of screen illumination, changing from the 500- to the 1000-watt lamp, 
seems to be greater than the increase of wattage, which is not in accord with 
usual practice. I am wondering whether the table does not involve an improved 
lens system, along with the higher- wattage lamp. That would tend to influence 
the costs and make them not strictly comparable. 

MR. MITCHELL:* Your assumption is correct that the efficiency of the optical 
system of the 1000-watt projector is greater than that of the optical system of 
other projectors that may have been used for this work. Attention is directed 
specifically to the fact that the same projector can use different lenses, so that 
its efficiency can vary as shown in Table I. 

MR. FARNHAM: The increase in screen illumination is seldom in accordance 
with the wattage. 

MR. KELLOGG: Do I understand, Mr. Dubray, that 48 interruptions per sec- 
ond was regarded as far as it was desirable or practicable to go for reducing 
flicker? 

MR. DUBRAY: Yes; that has been our practice. We project 16 frames per 
second and 3 interruptions per frame. The one-bladed shutter of the projector 
revolves three times while the picture frame is stationary. 

MR. KELLOGG: As I understand it, then, the interruption at the rate of 
three times a picture is desirable when projecting 16 frames a second, but at 24 

* Communicated . 



330 R. E. MITCHELL 

frames a second you hardly notice the benefit of three interruptions per picture, 
as compared with two. 

MR. DUBRAY: Two interruptions per second were used when sound came into 
existence for 35-mm. film. The linear speed of the projector was at that time 9 
feet per minute. For 16-mm. projection, the machine must be versatile, so that 
sound or silent pictures can be projected with it. The three interruptions per 
second are therefore necessary. 

It happens very often, also, that for analysis work a 16-mm. projector may be 
run at as low a speed as 10 picture frames per second, and an apparatus that can 
project without flicker at that speed is obviously very desirable. 

MR. SHAPIRO: In our experiments we have found at 48 interruptions per 
second the flicker becomes perceptible at a speed of less than 12 frames ; but at 24 
frames we can get along with about two-thirds of that number of interruptions 
per second without appreciable flicker at that speed. 

MR. FARNHAM: The paper stated that the brightness at the corners of the 
screen should not be more than 15 per cent less than the brightness at the center. 
I believe that you will find in actual experience that most projectors exceed that 
and it can be exceeded without serious effect, or without being seriously notice- 
able. The specification, probably, should be extended to take care of sudden 
changes between the corners. In other words, a 15 per cent change from the 
center of a picture to the corner is practically unnoticeable, whereas a 15 per cent 
change in a very short distance from the center becomes quite noticeable. The 
abruptness of the change is a factor, as well as the amount of the change. 

MR. DUBRAY: May I suggest, Mr. Farnham, that the recommendation made 
in the paper is not to take readings at the corners; but at the center, the top 
center and bottom center of the screen, and the left and right centers of the screen. 



STUDIO ACOUSTICS* 
M. RETTINGER** 

Summary. The possible means are discussed for obtaining the same amount of 
recorded reverberation with a directional microphone as is obtained when recording 
is done, in an equally configurated studio, with a non-directional microphone. 
First, there is assumed an equal distance between microphones and the source of 
sound; and, second, equal mean absorption in the two studios. 

The directional properties of the velocity ribbon and the uni- 
directional microphone have introduced a new and important factor 
in the theory of studio acoustics, namely, the imaginary solid cone of 
reception associated with the microphone. Mathematically, a solid 
angle is defined as the ratio of the surface of the portion of a sphere 
enclosed by the conical surface forming the angle to the square of the 
radius of the sphere. The unit is the steradian, and there are 4ir 
steradians to a sphere. 

For the sake of simplicity in the calculations it must be assumed 
that this cone of reception is clearly defined that is, that the only 
sound actuating the microphone lies within this cone, and that all 
sound without it is ineffective. While no such sharp boundary ac- 
tually obtains, a 100 per cent response existing only along the normal 
to the ribbon, the error introduced by regarding the solid cone as 
sharply limited is not very large, nor smaller nor larger for different 
frequencies, because the directional characteristic of the velocity mi- 
crophone is practically independent of the wavelength of sound. 

For a non-directional system of recording the energy density due 
to the direct sound at the microphone is 



E d = P/D^Cir (1) 

where P = power output, or rate of sound emission of the source ; assumed to be 

constant. 

D = distance between the microphone and the sound source. 
c = velocity of sound. 



* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Pacific Insulation Co., Los Angeles, Calif. 



331 



332 M. RETTINGER [j. s. M. p. E. 

Within a room having walls of average absorption a, the energy 
density due to reflected sound after steady-state conditions have set 



E r = 4P(1 - a)/cSa (2) 

where 5 = total surface in the room. 
Dividing (2) by (/), we get 

Er/E d = 16xZ) 2 (l - a)/aS (3) 

which ratio represents a measure of the received reverberation. 

In the case of the directional microphone, however, the reflected 
sound energy actuating the microphone is less, in the ratio of the solid 
cone of reception divided by 47r; thus, for steady-state conditions 

E' r = 4P(1 - a)K/cSak* (4) 



where K represents the solid cone of reception, expressed in steradians. 
Therefore, the amount of received reverberation becomes 

E'r /E d = 4> 2 (1 - a)/aS (5) 

which differs from equation (3) by the factor K/4ir ,which represents 
the difference in recorded reverberation when a directional micro- 
phone is used, provided the same studio and the same recording dis- 
tance are used. 

Assuming now, for the sake of illustration, two studios studio A 
and studio B which have the same shape and volume and in each 
one of which sound is recorded at the same distance from the source. 
Suppose in studio A that a condenser microphone is employed, and 
that the mean absorption in the studio is such as to permit good re- 
cording. The question arises as to what may be the mean absorption 
in studio B in which a directional microphone is used? Setting equa- 
tion (3) equal to equation (5), we may write 



- a)/aS = 4D*K(l - a'}/a'S 
or a' = aK/[aK + 4a-(l - a)} (6) 

where a' represents the mean absorption in studio B, in which the 
directional microphone is the same distance away from the source as 
the condenser microphone is in studio A. This mean absorption, 
a', to repeat, will make for the same amount of received reverberation 
in studio B as the mean absorption, a, makes in studio A . 



Oct., 1935] STUDIO ACOUSTICS 333 

The following table shows the relation between a and a' if K is 
taken equal to IT 



0.1 0.0276 

0.2 0.0555 

0.3 0.0970 

0.4 0.1430 

0.5 0.2000 

0.6 0.2730 

0.7 0.3690 

0.8 . 5000 

0.9 0.6820 

1.0 1 . 0000 



As was pointed out to the author by E. C. Wente, this table should 
be regarded with some caution. If the sound is picked up by a non- 
directional microphone within a room and the source is suddenly 
interrupted, there will be a drop in the generated voltage as soon as 
the direct sound has passed the microphone. After this the voltage, 
on the average, will decay logarithmically in accordance with the 
decay of the reverberant sound in the room. If now a directional 
microphone is used, there will similarly be an initial drop, somewhat 
greater, depending upon the value of K, after which the voltage will 
again decay logarithmically at the same rate as with the non-direc- 
tional microphone. This rate of decay can not be altered by a 
change in the directive properties of the microphone. If, however, 
the absorption in the room is increased, not only will the initial drop 
be greater, but the rate of decay of the subsequently developed volt- 
age will be greater. It is thus seen that the character of the 
voltage generated by a microphone when actuated by sound, such as 
speech or music, will not be quite the same when the ratio of E r to 
E d is reduced by making the microphone more directive as when this 
reduction is effected by an increase in the absorption within the 
room. 

This method of decreasing the mean absorption in order to obtain 
the same amount of received reverberation has the advantage of ob- 
taining a more uniform absorption characteristic in the studio, since 
materials having a comparatively small absorption coefficient at 
frequencies of 500 and 1000 cycles per second do not show such an 
abrupt decrease in their absorptivity at the low frequencies as do 



334 M. RETTINGER 

materials that are highly absorptive at the above-mentioned fre- 
quencies. 

There is another method of obtaining the same reverberation with 
a directional microphone that exists when recording is done with a 
non-directional microphone. It consists in lengthening the distance 
between the source and the microphone. Again setting equation 
(3) equal to equation (5), but assuming now the same mean absorp- 
tion in both studios, we get 

16ir> 2 (l - a}/aS = 4> 2 i(l - a)K/aS 
or D\ = -*D*/K (7) 



where D\ represents the distance between the directional micro- 
phone and the source of sound. 

Again assuming K equal to T, we get 

Di = 2D (8) 

which means that we can double the distance between the source and 
the directional microphone and still obtain the same amount of re- 
verberation that exists when recording is done with a non-directional 
microphone D units of length from the source. 

This method appears to be the more economical one, as the record- 
ing in the studio can be done with both the directional and the non- 
directional microphone without having to install variable absorbents. 
However, for scoring stages it may be more advisable to decrease the 
mean absorption in the stage when recording with a velocity micro- 
phone, because musicians, as a rule, can play better in live than in 
dead rooms. 

The author wishes to express his sincere appreciation to Professor 
V. O. Knudsen and Messrs. Townsend and Hansen of the Fox Film 
Corporation, whose interest in the author's work has been of great 
value to him. 

REFERENCE 

1 EYRING, C. F.: "The Reverberation Time in Dead Rooms," /. Acoust. Soc. 
Amer., 1 ( 1930), No. 2, p. 217. 



THE ARGENTOMETER AN APPARATUS FOR TESTING 
FOR SILVER IN A FIXING BATH* 



W. J. WEYERTS AND K. C. D. HICKMAN** 

Summary. The Argentometer consists essentially of a light-source, a glass cell 
for holding the solution to be tested, a photronic cell, and a microammeter. The 
action of the instrument depends upon the change of transparency of a solution 
containing silver and hypo after the addition of sodium sulfiide and certain other 
chemicals. Formulas for the three solutions required and complete instructions for 
making a silver determination are given. Estimations can be made on hypo solutions 
containing either potassium or chrome alum and the results can be read directly from 
a scale upon the microammeter. 

For many years the activity of fixing baths has been estimated by 
noting the time taken to clear a piece of unexposed film. When elec- 
trolytic silver recovery was introduced to the motion picture industry, 
this test proved insufficient and it became necessary to devise routine 
analyses for each of the more important constituents of the bath. 

The material which exerts the greatest influence upon both fixing 
and regeneration of the bath is the soluble silver salt derived from the 
material being processed. Many tests are available for the silver, but 
only the colorimetric sulfide reaction is quickly and easily applied. 
At its simplest, a sample of the bath is diluted with water and treated 
with a little sodium sulfide solution. A brown precipitate or color is 
produced, the depth of which is roughly dependent upon the silver 
present. The quantity is estimated by comparing the color of the test 
solution, held before an illuminated screen in a glass tube, with other 
tubes containing a known concentration of silver sulfide. 

There are two serious drawbacks to this procedure. The standard- 
ized comparison solutions alter in color in an unpredictable manner 
and become unreliable after the second month. The color developed 
from the solutions under test varies not only with the quantity of sil- 
ver, but with the acidity, the sulfite and alum content, the quantity 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. Communica- 
tion No. 548 from the Kodak Research Laboratories. 
** Eastman Kodak Co., Rochester, N. Y. 

335 



336 W. J. WEYERTS AND K. C. D. HICKMAN [J. S. M. P. E. 

of gelatin and stale developer, as well as the metallic salts from the 
tanks and pipes, accumulating in the bath. 

The method of silver estimation presented here dispenses with the 
silver standards and minimizes or eliminates the disturbing chemical 
factors. The method comprises (1) a photoelectric comparator, i. e., 
the "Argentometer" ; (2) a precipitation procedure employing buffered 
solutions. 

THE ARGENTOMETER 

The silver comparator illustrated in Figs. 1 and 2 is a hollow metal 
box which contains a lamp A, a transparent vessel B for the liquid 




FIG. 1. The Argentometer, complete. 

under examination, and a photronic cell C. The box is closed with a 
sloping lid (Fig. 2), to one side of which is inserted a microammeter D, 
scaled in grams of silver per liter (Fig. 3) . The lid forms a convenient 
rest for pencil and notebook. If there is any point of novelty in this 
simple instrument, it is that the need for a standard lamp and regu- 
lated voltage is obviated. The solution under test is diluted with 
water and placed into the glass vessel, which is then pushed into the 
box. The current is switched on and the lamp is moved forward or 
backward upon its adjustable slide E (Fig. 1), until a standard deflec- 
tion of 150 microamperes is registered upon the lower scale of the 
meter. This deflection is adopted as the measure for zero content of 
silver, and is so marked upon the upper silver scale. The glass vessel 



Oct., 1935] THE ARGENTOMETER 337 

is now withdrawn and the precipitating solutions added, stirring; 
after which the vessel is again thrust into the box. The meter needle 
recedes to a lesser deflection, which corresponds to the depth of color 
generated in the solution by the interaction of the reagent and the 
silver salts. The quantity of silver present is read directly upon the 
upper scale, which is calibrated backward from the place of maximum 
deflection. 

It will be noted in Fig. 3 that the scale is most open and the sensi- 
tivity is greatest for solutions poorest in silver a very practical ad- 
vantage. The establishment of the zero point at the place of maxi- 




FIG. 2. Showing the three important elements of the Argentom- 
eter: (A) the light-source; (B) the glass cell for holding the solu- 
tion; (C) a photronic cell; and (D) a microammeter. 

mum deflection renders the instrument relatively unaffected by chang- 
ing conditions. Should the lamp burn out or should it be necessary 
to change to another voltage, a new lamp may be inserted and the 
appropriate position found upon the lamp slide. Should the test solu- 
tion be dark colored or cloudy instead of colorless and transparent, the 
opacity is compensated by the lamp setting, and only the additional 
opacity conferred later by the silver sulfide is registered. 

THE SOLUTIONS 

The opacity of a given concentration of silver sulfide depends upon 
the covering power of the colloidal sulfide particles, and this in turn 
depends upon the pH, the neutral salt content, and the "protective" 
powers of the solution. In fixing baths containing hypo, alum, and, 



338 



W. J. WEYERTS AND K. C. D. HICKMAN [J. S. M. p. E. 




FIG. 3. The microammeter 
is calibrated directly in grams 
of silver per liter. 



perhaps, iron, the acidity should be high enough to prevent precipita- 
tion of alumina and chromic hydrate or ferrous sulfide, and sufficiently 
low to prevent the separation of colloidal sulfur. Citric acid and 

sodium citrate are used to keep the 
pH in the proper range, and it is 
satisfactory to find that the addition 
of a standard quantity of these 
reagents will adjust fixing bath 
samples of widely different composi- 
tion and condition. The "protec- 
tive" power of a fixing bath varies 
enormously with its age, and it is 
thus necessary to add sufficient gela- 
tin to the test solution to give a large 
excess over any likely to be present. 
A plain sodium sulfide solution has been 
found unsatisfactory as a precipitant 
because it may generate colloidal sulfur upon meeting the acidified 
sample. Accordingly, sodium sulfite is added to the stock solution 
of this reagent. 

OPERATION 

The action of the estimator depends upon the change of trans- 
parency of a solution containing silver and hypo after the addition of 
certain chemicals. These are made up as in Table I to form stock 
solutions A, B, and C: 

TABLE I 

Solution A : Metric Avoirdupois 

Citric Acid 9 gm. l l / t oz. 

Sodium Citrate 100 gm. 13 oz. 

Water to make 1000 cc. 1 gal. 
Solution B: 

Gelatin 4 gm. */2 oz. 

Water to make 1000 cc. 1 gal. 

Clove Oil (few drops for preservation) 

Soak the gelatin in a small amount of water until well swollen; then heat 
gently, stirring, until all is in solution; dilute to proper volume with 
water, pour into a bottle, add the clove oil, and shake a few times. 
Solution C: 

Sodium Sulfide (C. P. crystals) 10 gm. 1 oz. 

Sodium Sulfite (anhydrous)* 6 gm. 3 /4 oz. 

Water to make 100 cc. 10 oz. 



Sodium bisulfite or metabisulfite must on no account be used. 



Oct., 1935] THE ARGENTOMETER 339 

PROCEDURE 

The instrument as supplied is fitted with a 110- volt, 40- watt, 
frosted bulb lamp. The plug at the side of the box is therefore con- 
nected to a 110-volt source. Should the house-supply differ from 110 
volts, the lamp should be changed to a 40-watt, frosted bulb of the 
appropriate voltage. 

To make an estimation, a sample of the hypo solution is measured 
out with a 2-cc. pipette and allowed to drain into a 50-cc. graduated 
cylinder. Five cubic centimeters of solution A are added, and then 
5 cc. of solution B; after which the cylinder is filled to the 50-cc. 
mark with water. The contents are now transferred to the glass cell, 
which is placed into the sliding carriage and pushed into the estimator 
box as far as it will go. This will locate the cell in position between 
the light-source and the photronic cell. The current is switched on 
and the light moved backward or forward, until the needle of the 
microammeter points to zero upon the upper silver scale (150 micro- 
amperes upon the lower). The cell is pulled out and 1 cc. of the 
sodium sulfide solution C added from a pipette. The solution is 
stirred with a glass rod, the end of which is protected by a short 
piece of rubber tubing, and the cell is pushed back into the box. 
The microammeter needle will assume a new position from which the 
silver content of the sample can be read upon the upper scale. 

It occasionally happens that the current supply for the lamp fluc- 
tuates so badly that the zero setting alters during an estimation, giv- 
ing faulty results. When this is suspected, or when great accuracy is 
required, it is best to take a check-zero reading as follows : After the 
sample has been mixed with solutions A and B and the additional 
water, and placed into the cell, the latter is pushed into the instrument 
as before, and a careful zero adjustment made. The cell is with- 
drawn and the position of the needle upon the microampere scale is 
noted. This is the check-zero, to which the lamp can be adjusted, 
should the current alter during subsequent proceedings. It is well to 
find the zero and the check-zero two or three times in succession. The 
sodium sulfide solution can now be added and the estimation carried 
out as before. Immediately before and after taking the reading for the 
silver, the cell should be withdrawn to see that the needle points ex- 
actly to the check-zero. If it does not, the position of the lamp should 
be adjusted. 

If the glasses of the cell become broken, they can be replaced by 
new ones which need not be accurately of the same kind or thickness 



340 W. J. WEYERTS AND K. C. D. HICKMAN 

as the broken ones. It is, however, essential that the plates be ad-' 
justed to remain exactly 5 / 8 of an inch apart. The spacing can be 
adjusted by inserting shims between the edges of the glass plates and 
the brass frames. 

SCOPE OF OPERATION 

The estimation may be performed upon hypo baths containing 
either potassium or chrome alum, since the color of the chrome alum 
is compensated for by moving the light until the needle indicates 
zero before the sodium sulfide is added. 

Concentrations of silver as high as 3 grams per liter are often en- 
countered in commercial fixing baths for prints, and 8 to 10 grams per 
liter for films. Proper fixation can not be attained with such over- 
worked solutions, traces of silver being left in the paper and in the 
emulsion. This ultimately causes yellowing of prints and the growth 
of a brown surface sheen upon negatives or motion picture film. Com- 
plete removal of silver is desirable for permanence, and is imperative 
if the materials are to be toned or dyed. The safe upper limit for the 
silver content of print fixing baths may be placed at I 1 / \ grams per 
liter, with 4 grams as the upper limit for films. Desirable working 
concentrations are 0.5 gram and 1.5 grams per liter, respectively. 



REPORT OF THE PROJECTION PRACTICE COMMITTEE* 
PROJECTION ROOM PLANNING 



The following recommendations for projection room planning have 
been formulated after an exhaustive study by the Committee and are 
submitted for adoption as standards. The Committee urgently 
recommends their acceptance by all architects and builders in con- 
structing and remodeling projection rooms so that a greater uni- 
formity of projection room construction will exist in the future. 

In following these recommendations the proper authorities 
should in all cases be consulted for possible deviations therefrom. 
Any fire protection requirements specified herein are in accordance 
with the regulations of the National Board of Fire Underwriters. 
However, these requirements are neither complete nor in detail, and 
it is the plan of the Committee to work with the Fire Underwriters in 
the near future in the preparation of a comprehensive set of recom- 
mendations for adoption as standard regulations by the industry. 

PROJECTION ROOM CONSTRUCTION 

General. Three layouts are presented, viz., Figs. 2, 3, and 4, which 
were planned with careful regard for flexibility, simplicity of construc- 
tion, and ease of operation. The particular plan to be followed 
should be selected according to the size of the theater and the manner 
of operating it. The key to the symbols used on the plans is shown in 
Fig. 1. 

The projection room shall be fire-proof and sound-proof, and all 
walls exposed to the theater shall be of tile, brick, gypsum, or other 
approved fire-resisting material. It shall have a minimum height of 
10 feet and a maximum of 12 feet. The minimum depth shall be 12 
feet. The length of the projection room shall be governed by the 
quantity and the kind of equipment, as shown in the plans and in 
accordance with local requirements. Consideration should always be 
given for probable future needs. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 

341 



342 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



The Committee recommends that the projection room be located 
outside the fire-wall of the theater, and that it be so situated that the 
projection angle shall not exceed 15 degrees. 

Floor. The floor of the projection room shall be sufficiently 
strong and solid for the load it is to bear, and shall be constructed in 
accordance with local building regulations. A generous factor of 
safety should be allowed. 

A type of floor construction that is recommended consists of (1) a 
reenforced concrete floor-slab not less than 4 inches thick; (2) a 
tamped cinder fill above the floor-slab not less than 2 inches thick; 



TABLE I 


-j 




Splay of Projection 


and Observation Ports 


s 




Thickness (t) \ 
of Wall (Inches) | 4 


6 


8 


10 


12 


14 


16 


Angle of 














\ 


Projection (6) 






Splay 


(Inches) 






\. ^ 


5 


3 /8 


V 


3 /4 


7 /8 


11/8 


! 3 /8 




10 


3 /4 


1 


P/8 


1V8 


2 


2 X /4 


2 5 A 


*N- ,-, 


15 


iVs 


! 5 /8 


2Va 


2 3 / 4 


3V4 


3 3 / 4 


4V4 


\. If 


20 


iVi 


2 X /4 


2 7 / 8 


3 5 /8 


4 3 / 8 


6V. 


5 7 / 8 


^^ 


1__ 


25 


! 3 /4 


2 7 / 8 


3 5 /8 


4 7 /8 


5 7 / 8 


6 7 / 8 


7 3 /4 






30 


2 3 / 8 


3Vi 


4 3 / 8 


5 7 / 8 


TVi 


8V4 


9Vi 


f-* 





and (5) a trowelled cement finish above the cinder fill not less than 2 
inches thick. 

Ceiling. The ceiling shall be of plaster or cement suspended on 
metal laths or other suitable material. 

Walls. The finished walls of the projection room shall be not less 
than 6 inches thick, including an inside and an outside layer of plaster 
at least 3 / 4 inch thick. In all cases, the inside surface of the front wall 
shall be smooth and without structural projections. 

Acoustic Treatment. The inside walls and ceiling of the projection 
room shall be finished with an acoustic plaster or other sound-absorb- 
ing material approved by the proper authorities. 

Projector Ports. The finished projector ports shall be 10 inches 
wide and 12 inches high (Fig. 5). The bottom of the opening shall be 
splayed in accordance with Table I. 

Tables II and III apply to certain well-known makes of projectors. 
Table II gives the distance from the front wall to the center of the 
conduit outlets in the floor for the projectors. Table III gives the dis- 



Oct., 1935] PROJECTION PRACTICE COMMITTEE 343 



^_ Ceiling Outlet "Reelite" 

-(^- Wall Bracket 

-<)- Ceiling Outlet Canopy Switch Type 

-<o)- Ceiling Outlet 

(0j Outlet in Floor for Dissolving Stereopticon 

[7J Outlet in Floor for Effect Machine 

Q Outlet in Floor for Flood Lamp 

[3 Outlet in Floor for Projector Arc 

El Push Button 

|oSJ Double Baseboard Receptacle 

15 M.P.M. Motor Outlet 

rt House Phone 

Wall Switch for Ceiling Lights 



Wire Sizes 

Low-Intensity 30 A. No. 4 

Reflector High-Intensity 75 A. No. 2 

High-Intensity 125 A. No. 00 

Super High-Intensity 200 A. 200,000 C. M. 



FIG. 1. Key of symbols for projection room layouts. 



344 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



- a 

Q.8 



J?T}. OT^(MO<M^COOOO(N^COOOOC<1 

S| I. ^^^^.H^M 



T^ CD 
<N <M 



Si 



cococococococococococo 
cococococococococococo 



o o o o o o 
co co co co co co 



CO CO CO CO 



is >> 



a a:: 



^s 

1 



i^^-o 
! o a 



l 



Oct., 1935] PROJECTION PRACTICE COMMITTEE 345 

tance from the floor to the center line of the projector ports for differ- 
ent angles of projection. 

In preparing this report, the Committee encountered considerable 
difficulty in forming suitable recommendations for the location of the 
projector ports. This was due to the non-uniform design of the vari- 
ous makes of projectors. Table IV is a tabulation of two constants 
for various angles of projection which, when substituted in the formula, 
will give the distances A and B for makes of projectors other than 
those included in Tables II and III. 

The Committee recommends the use of means other than glass in 
projector ports to prevent transmission of noise from the projection 
room to the auditorium, such as reducing the free aperture of the port 
to the minimum essential for projection. 

Observation Ports. The free aperture of the observation ports shall 
be 12 inches wide and 14 inches high, and the distance from the floor 
to the center line of the openings shall be in accordance with Table V. 
The bottom of the port shall be splayed in accordance with Table I. 

The observation ports shall be fitted with a good grade of plate 
glass set at an angle as shown in Fig. 6, and provided with a rubber 
frame between the glass and the sides of the port hole in order to 
reduce the transmission of sound from the projection room into the 
auditorium. The glass shall be hinged at the centers of the side edges 
so that by swinging it to a horizontal position, both sides can be 
cleaned from the projection room. 

Other Ports. All other ports, such as those intended for effect pro- 
jectors, dissolving stereopticons, or single spot-lamps, shall be 30 
inches wide and 36 inches high. The distance from the floor to the 
center line of the ports shall be 38 inches. The minimum spacing 
allowed between these ports shall be as shown upon the plans. The 
bottom of the ports shall be splayed in accordance with Table I. The 
placing of these ports to the right or the left of the projectors shall be 
optional and according to conditions. 

Floor Covering. Where local regulations permit, the floor of the 
projection room should be covered with a good grade of fire-proof 
material; otherwise, the cement should be painted or filled. The floor 
covering should be laid before the equipment is installed. The floors 
of rooms adjacent to the projection room should be painted with a 
good grade of paint for concrete. 

Projection Room Painting. The color of the projection room walls 
and doors shall be olive green to the height of the door lines. Acoustic 



346 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



material should either be painted in accordance with the instructions 
of the manufacturer of the material, or materials of the specified colors 
should be chosen : The walls above the door line and the ceiling shall 
be a buff color. All iron work of projection ports shall be covered 
with at least two coats of flat black paint. All other rooms shall be 
painted buff. 
Projection Room Lighting. An individual approved ceiling fixture 

TABLE IV 

Method of Locating Projector Port for Any Projector 
h = H + rA - DB 

Projection 

Angle 
(Degrees) A B 





2 

4 

6 

8 

10 

12 

14 

16 

18 

20 

22 

24 

26 

28 

30 



1.00 
1.00 
1.00 
1.01 
1.01 
1.02 
1.02 
1.03 
1.04 
1.05 
1.06 
1.08 
1.09 



1.16 



0.00 
0.04 
0.07 
0.11 
0.14 
0.18 
0.21 
0.25 
0.29 
0.33 
0.36 
0.40 
0.45 
0.49 
0.53 
0.58 




] 




<+^ PIVOT 


r 


h 


\ 


1 

FLOOtZ. 


- ^ 

1 1 

Q 


^ 
\ 

i 


//////////////////'//'A 



H is the height of the center of the projector pivot from the floor ; r is the radial 
distance of the optical center line above the center of the pivot ; D is the distance 
of the center of the pivot from the front wall of the projection room ; </> is the angle 
of projection; and h is the required height of the center of the port from the floor 
of the projection room. Select the values of . a and b corresponding to the angle of 
projection, and substitute in the formula. 

with canopy switch shall be installed for each piece of equipment, and 
shall be placed in line parallel to the front wall at a distance not less 
than 18 inches nor more than 24 inches from the front wall. The outlet 
connected to the emergency lighting system shall be located in the 
ceiling midway between the extreme ends of the projection room and 
4 feet from the back wall. Small projection rooms shall be equipped 
with one approved "reelite," and large projection rooms with two 
such lights conveniently located. 



Oct., 1935] 



PROJECTION PRACTICE COMMITTEE 



347 



Conduits. (a) Conduits shall in all cases be concealed, and all 
boxes shall be of the flush-mounting type, (b) The size of conduits 
for projection arcs shall be in accordance with the wire sizes indicated 
in Fig. 1, and in conformance with the regulations of the proper 




FIG. 5. Standard projection port construction. 

authorities. These sizes anticipate the need for increased capacity, 
and should be adhered to in order to provide space for pulling in 
larger wires as needed, (c) Conduit for sound equipment shall con- 
form to the type of sound equipment to be installed. The manufac- 
turers of such equipment should be consulted with regard to the 



348 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



proper layout of the sound system before proceeding with the in- 
stallation. 

Projection Room Heating. Proper provision shall be made for heat- 
ing the projection room. The same facilities used for heating the 
theater should be extended to the projection room. 

Projection Room Ventilation. An exhaust system of ample capacity 
shall be provided for the projection room and other adjacent rooms 
used in connection with projection equipment. All arcs of whatever 
description shall be connected into the ducts of a separate exhaust 
system containing a blower type of exhaust fan. 

There shall also be a separate opening or vent flue in the main pro- 
jection room leading directly to the nearest outside air. Such flues 

TABLE v 

Height of Observation Port 
(See Figs. 2, 3, and 4) 



Proj. Angle 
(Degrees) 


Distance, Floor 
to Center of 
Observation Port Proj. Angle 
(Inches) (Degrees) 


Distance, Floor 
to Center of 
Observation Port 
(Inches) 


-6 


51 


+10 


43 


-4 


50 


+12 


42 


-2 


49 


+ 14 


41 





48 


+ 16 


39 


+2 


47 


+ 18 


38 


+4 


46 


+20 


37 


+6 


45 


+22 


37 


+8 


44 


+24 


37 






+26 


36 



shall be at least 78 square inches in cross-section and constructed of 
incombustible materials. When the projection room is in use, a cur- 
rent of air shall be maintained through the room to the outside air at 
a minimum rate of 50 cubic feet a minute, and sufficient to furnish a 
complete change of air in 10 minutes. In cases where the theater is 
air-conditioned, the projection room shall be connected into the main 
duct of this system. 

Additional Rooms. A separate room shall be provided solely for 
the rheostat equipment. This room shall be provided with ventilat- 
ing means as previously set forth. An additional and separate room, 
properly ventilated, shall be provided for the projection arc supply 
equipment. Where local regulations require, a properly ventilated 
room should be provided for rewinding. 



Oct., 1935] 



PROJECTION PRACTICE COMMITTEE 



349 



Toilet and Washrooms. Hot and cold water and other toilet facili- 
ties shall be installed, and located convenient to the projection room. 
Suitable space shall also be provided for clothes lockers. 

PROJECTION ROOM EQUIPMENT 

Projectors and Spacing. Where two projectors are used, they shall 
be equally spaced upon either side of the center line of the auditorium. 




-J 



FIG. 6. Standard observation port construction. 

When three projectors are used, the center projector shall be placed 
upon the center line of the auditorium. The distance between pro- 
jectors shall be 4 l /2 feet, measured between lens centers, for projection 
distances greater than 100 feet. For projection distances less than 
100 feet, the spacing shall be 4 feet. 



350 



PROJECTION PRACTICE COMMITTEE [j. s. M. p. E. 



/-f/NISH PLASTER *^-^^ 



NOTE 

ne SHUTTEG OPEW MIGHT 5f o EXCEEDS WEIGHT c 

KCfDS ^S/^HT C 




PULLCY DETAIL 



WSL 



/*' '4 FLAT 
SANDlKOfj 



MH 



~~ 



e 



PULLEY HOUS/N<J 



SECT/ON A-A 




FIG. 7. Standard projection port shutters. 



Oct., 1935] PROJECTION PRACTICE COMMITTEE 351 

Projection Arc Supply and Location. In those cases where the pro- 
jection arc supply consists of machinery that generates acoustical 
hum or n^hanical vibration, the use of acoustical or mechanical in- 
sulation will be required. Rotating machinery used for projection 
arc supply shall be located as remotely as possible from the auditorium 
and the projection room. Arc supplies other than rotating equipment 
may be located in the room adjacent to the projection room, taking 
precautions to place it at least 4 feet away from the sound equipment. 

Power Supply to Equipment. Where line-voltage variations are 
greater than =t 3 per cent, the power company should be requested to 
rectify the condition. In those cases where it is impossible to main- 
tain a steady line supply into the theater, either manually controlled 
or automatic regulators should be installed. 

Film Storage. Approved film storage cabinets having a capacity 
sufficient to accommodate all the film in use in the theater at any one 
time shall be installed. The film shall be kept in such cabinets at all 
times except when being projected or rewound. Any film in addition 
to that being used for the current show or in excess of that permitted 
by local authorities, shall be kept in original shipping containers. 

FIRE PROTECTION 

Projection Port Shutters. (Fig. 7.) These shall be constructed of 
iron guides not thinner than 16-gauge, built up of iron flats, 2 inches 
wide and l / 8 inch thick, with spacers 1 inch wide and J / 4 inch thick, in 
which the shutter may slide. The shutter shall be made of not less 
than 10-gauge iron, or of other approved fire-proof material. The 
bottom sill of shutter tracks shall be provided with leather bumpers. 

One type of mechanism for the shutter system having the approval 
of the Committee consists of a suitable rod which may be constructed 
of I 1 /z-inch pipe mounted upon the front wall of the projection room 
in a series of ball-bearing brackets in such a manner that the rod may 
revolve freely in the bearings. 

The shutter system shall be located a sufficient distance below the 
ceiling line to admit of easy operation. At each port, and securely 
fastened thereto, shall be a chain or rod of approved design attached 
to a metal ring fitting loosely over a pin inserted into the rod about 
45 degrees upward from the horizontal, so that the revolving of the 
rod shall cause the pin to fall to a down-vertical position and permit 
the ring to slip off and drop the shutter. 

Into each shutter cord or chain shall be inserted an approved 



352 PROJECTION PRACTICE COMMITTEE 

fusible link. The master control cord shall be so arranged through a 
system of pulleys in conjunction with a counterweight that either 
automatic or manual operation will permit the shutters to drop. The 
master manual control cord shall be located at each of the entrances 
of the projection room. In addition, the master control cord shall 
be furnished with fusible links placed approximately 10 or 12 inches 
above and immediately upon the center line of the projector maga- 
zine. All larger shutters shall be provided with an additional counter- 
weight (Fig. 7), to facilitate manual operation of these shutters. 

Emergency Exhaust Fan. An exhaust fan of sufficient capacity to 
remove all smoke and gas in case of fire shall be provided, and this fan 
shall be so connected to the port shutter controls that its full capacity 
will be automatically made available upon the dropping of the shut- 
ters. The fan and air duct shall be of fire-proof construction and lo- 
cated in accordance with local requirements. 

Doors. The doors shall be an approved metal or metal-clad type, 
swinging outward from the projection room, and shall be provided 
with door-checks or other approved door-closing devices. 

Exits. Exits shall be provided strictly in accordance with local 
authorities having jurisdiction, particularly with reference to size and 
location. It is recommended that never less than two exits from the 
projection room be provided. At least one of these exits should be of 
the conventional stairway type, with risers not in excess of 8 inches 
and a minimum tread to each step of not less than 9 inches, and of 
sufficient size to permit transportation of equipment. 

It is recommended that the secondary exit provide a means of 
access to the ground, which may be via a roof scuttle or doorway ac- 
cessible from the projection room by means of steps or a ladder. 

Windows. Where a projection room is built against the exterior 
wall of a structure, one or more windows should be provided in it. 

Fire Extinguisher Equipment. The local authorities having juris- 
diction should be consulted regarding the proper type, number, and 
location of fire extinguishing equipment. 

J. O. BAKER, Chairman 

J. O. AALBERG J. J. FINN P. A. McGuiRE 

T. C. BARROWS E. R. GEIB R. MIEHLING 

F. C. CAHILL, JR. H. GRIFFIN M. D. O'BRIEN 
J. R. CAMERON S. W. HANDLEY H. RUBIN 

G. C. EDWARDS J. J. HOPKINS J. S. WARD 

J. K. ELDERKIN C. F. HORSTMAN V. A. WELMAN 



REPORT OF THE SOUND COMMITTEE 



The Sound Committee this year has decided to concentrate upon 
four projects, the first two of which will be discussed in some detail 
because the Sound Committee will need the cooperation of every one 
if it is to accomplish anything on them. The object of the first proj- 
ect is to achieve greater uniformity in the sound records from the 
various producers. Certainly, it will be agreed that that is necessary 
and desirable; but at once the question arises as to what is the proper 
or ideal recording characteristic. The Sound Committee three years 
ago recommended 1 to the Standards Committee that the dividing 
line between recording and reproduction should be the release print 
that all losses incurred up to the release print" . . . should be compen- 
sated for in the recording operation. The frequency characteristic 
of the reproducing apparatus should be flat except for a correction for 
whatever slit is used." This recommendation, however, has not 
been adopted, as yet, by the Standards Committee, and it is the sense 
of the present Sound Committee that the recommendation should be 
reconsidered in the light of the developments and data accumulated 
during the past three years. 

Before any conclusions can be reached on this point it seems neces- 
sary to obtain data on the recording frequency characteristics used 
by the various studios at present. This, in turn, raises the question 
of how to obtain comparative data from the various studios. It 
appeared from a discussion of the question that a great deal of skill, 
patience, and special calibrated precision equipment are required if 
satisfactory results are to be obtained. To determine whether meas- 
urements made by different organizations would check, a frequency 
film was made and the same print was measured by five different 
organizations. When the results were plotted upon the same sheet 
of paper they revealed excessive deviations at frequencies higher than 
3000 cycles and smaller ones at frequencies lower than 200 cycles. 
Even the results from two of the most skilled research organizations 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 

353 



354 SOUND COMMITTEE fj. S. M. p. E. 

in the country deviated by amounts that represented the limit in 
experimental error. 

It was decided, therefore, that a much simpler method of obtaining 
comparative data from the various studios would be to establish a 
"Frequency Reference Standard" in the form of a carefully prepared 
frequency film. In such a film the actual level that is recorded is of 
little importance. For comparative purposes it is necessary only 
that the various organizations concerned measure the reproduction 
from this Reference Standard upon any machine and any system that 
they may happen to be using; and then to measure, under identical 
conditions, a film record made by themselves, and to report the de- 
viation from this Reference Standard. If two organizations obtain 
the same deviation it is obvious that their recording characteristics 
will be the same ; and if different, a comparison of the deviations will 
show the location and magnitude of the differences. 

Such a film has been made for the Sound Committee through the 
courtesy of the RCA Manufacturing Company, and the master nega- 
tive is in the hands of the Chairman of this Committee. The No. 1 
print has been designated the "Primary Frequency Reference Stand- 
ard," and the Eastman Kodak Company has kindly offered to meas- 
ure its frequency characteristic once each year on their microdensi- 
tometer, to determine the stability of this film as a reference standard. 

It is planned to hold the number of prints made from the master 
negative to an absolute minimum. Each succeeding print made 
from the master negative, beginning with the No. 2 print, will be des- 
ignated a "Secondary Frequency Reference Standard," and a calibra- 
tion will be provided with it which will show the deviation between it 
and the Primary Frequency Reference Standard. The secondary 
standards should not be used as test reels, but should be used only 
for the purpose of calibrating test reels. At intervals indicated by 
experience, the secondary standards can be calibrated in terms of the 
primary standard. Thus we shall have for the work of this Commit- 
tee, a datum plane or bench-mark to which all measurements can be 
referred and, therefore, correlated with each other. To begin with, 
it was decided to make six secondary frequency reference standards. 
The stock was furnished through the courtesy of the Eastman Kodak 
Company, the printing and developing through the courtesty of De 
Luxe Laboratories, and the calibrating through the courtesy of 
Warner Brothers. 

The film referred to above has the variable- width type of track. 



Oct., 1935] SOUND COMMITTEE 355 

A second similar film of the variable-density type is being made for 
the Committee so that the relative permanency of the two types of 
track as reference standards can be studied. 

Within the next few months, therefore, the various studios may 
expect to be requested to furnish to the Sound Committee informa- 
tion on their recording characteristics, expressed in terms of their de- 
viation from the primary frequency reference standard. From these 
data it is hoped that the Sound Committee can arrive at a recording 
frequency characteristic that they can recommend. 

The object of the second project is to obtain, from the level stand- 
point, a more artistic rendition of sound in the theater. In other 
words, the "level balance" that the producer had in mind at the 
time the record was made should be preserved, and the projectionist 
should be enabled to pre-set his reproducing equipment, without 
rehearsal, so as to create the effect intended. 

If all producers will determine the normal or reference setting for 
reproduction in their review rooms, and all theaters will do the same, 
it will then be possible to mark upon all release prints the deviation 
from this normal which would be required in any review room or 
theater to create the effect and loudness desired. In order to deter- 
mine the normal or reference setting in review rooms or theaters, 
taking into account also personal tastes as to loudness, a standard 
test reel will have to be selected. This may be the SMPE standard 
test reel or a reel chosen especially for the purpose by the Committee. 

Two factors determine the relative loudness, on a given equipment, 
of two successive reels: (1) the percentage of modulation used in 
making the negative, and (2) the processing of the print, in which 
density or transmission is the most important factor. If the pro- 
ducer, when previewing the No. 1 print made from the release nega- 
tive, would determine the deviation from his normal setting required 
to create the loudness and the effect he wishes, this deviation ex- 
pressed in decibels would represent the deviation from normal in any 
other place for all prints having the same density as the one that he 
previewed. If the laboratory making the rest of the release prints 
were given this information, as to the level in decibels above or below 
normal at which prints having the same density as the No. 1 print 
should be played, the laboratory could then determine from a table 
showing the variation of level with density the correct playing level 
for all prints whose densities differ from that of the No. 1 print. This 
information could then be written or punched at a suitable place 



356 SOUND COMMITTEE [J. S. M. p. E. 

upon the leader; for example, it might be incorporated as part of the 
24 frames of "identification leader" of the standard leader of the Acad- 
emy of Motion Picture Arts and Sciences. After noting the devia- 
tion from normal marked upon the reel, the projectionist could play 
the reel at the correct level in his theater by making the indicated 
setting on his equipment. 

The object of the third project is to study the processing charac- 
teristics for sound-film used by various producers, to the end that 
greater uniformity may be achieved in processing methods. As the 
Technicians' Branch of the Academy already has a Committee work- 
ing upon the problem, it is felt that the Sound Committee of the 
S. M. P. E. should endeavor to cooperate in any way possible with the 
Academy in an effort to expedite the work. We believe that when 
the Academy first undertakes to standardize the measurement of film 
densities it is attacking the problem in the correct manner. The 
Sound Committee is being urged to study the problem in the hope of 
assisting the Academy to eliminate the rather wide divergence 
apparently existing at the present time in the measurement of den- 
sities by the various studios. 

The object of the fourth project is to evolve, if possible, a simple 
and inexpensive method of making rough measurements of the acous- 
tic frequency characteristic of review rooms and theaters, to the end 
that review rooms and theaters may be adjusted so that a given 
print will sound more nearly the same when reproduced in the various 
places. It has been suggested that if a wobble-frequency film, ful- 
filling certain conditions, were available, measurements could be made 
quickly and inexpensively, enabling the engineer more nearly to ap- 
proximate ideal reproduction conditions. This proposal was not 
made with the thought in mind that it would supersede the more 
refined and accurate work now being carried on by the acoustical 
experts, but would make available to those who could not afford or 
who did not have access to the more refined and accurate methods a 
means for making a first or coarse adjustment of their reproduction 
conditions. 

In closing, the Committee wishes to commend the newsreels upon 
the marked success that they have attained in standardizing the loud- 
ness of their release prints. 2 



Oct., 1935] SOUND COMMITTEE 357 

P. H. EVANS, Chairman 
C. DREHER, Vice- Chairman 

M. C. BATSEL W. C. MILLER O. SANDVIK 

R. M. EVANS K. F. MORGAN E. I. SPONABLE 

L. G. GRIGNON W. A. MUELLER R. O. STROCK 

E. H. HANSEN L. L. RYDER S. W. WOLF 

J. P. LIVADARY W. WOLF 

REFERENCES 

1 Report of the Sound Committee, /. Soc. Mot. Pict. Eng., XIX (Aug., 1932), 
No. 2, p. 166. 

2 BATTLE, J. A.: "Improvements in Sound Quality of Newsreels," /. Soc. 
Mot. Pict. Eng., XXV (Aug., 1935), No. 2, p. 154. 

DISCUSSION 

MR. SANDVIK: The Sound Committee feels that the question of uniform fre- 
quency characteristic of the talking motion picture, as heard in the theaters, 
is the most important problem that it has at hand, and we should welcome 
very much any information available bearing upon the problem. It is a very 
far-reaching problem, taking in all operations that have any relation to the sound, 
from the recording studio to the theater (i. e., the acoustics of the theater), 
and all that the sound Committee can hope to do is to initiate thoughts along these 
lines, hoping that finally something will come out of them. 

For the purpose of illustration let it be assumed that through a careful study of 
the theater acoustics, and other factors that enter into consideration, the most 
satisfactory characteristic of the frequency spectrum that should be fed into the 
average theater has been decided upon. Let it further be taken for granted that 
somewhere in the process it will be necessary to equalize; that is, to boost the 
high-frequency end of the spectrum with respect to the low-frequency end. 
The question then is, at what stage or stages in the process should the equaliza- 
tion take place? Or, to make it more concrete, which will result in a greater 
volume range, for example, to introduce equalization in the original recording 
system or in the final reproducing system? The answer to this question is quite 
evident, but the information in the literature- on this general subject is very 
limited. 



SYMPOSIUM ON NEW MOTION PICTURE APPARATUS 



During the Spring Convention at Hollywood, Calif., May 20-24, 1935, a sympo- 
sium on new motion picture apparatus was held, in which various manufacturers of 
equipment described and demonstrated their new products and developments. Some 
of this equipment is described in the following; the remainder will be published in 
subsequent issues of the Journal. 

OZAPHANE FILM AND THE CINELUX PROJECTOR* 

A. M. CHEFTEL** 



The exhibition of motion picture film in suburban areas in France and in the 
colonial possessions is carried on under circumstances requiring the greatest 
possible economy in theater equipment, release print cost, and distribution ex- 
penses. 

It is believed that sufficiently economical operation is made possible through 
the use of the very thin, low-cost, slow-burning Ozaphane film for release prints. 
Ozaphane film is made from a cellulosic sheet material 0.04-0.05 millimeter 
(about 0.002 inch) thick, containing the material from which a photographic 
image is formed by the diazotype process. Thus, an additional image-carrying 
layer is unnecessary, and the weight and thickness of the material is a minimum. 
A reel that holds 300 meters (984 ft.) of regular motion picture film will hold 
1000 meters (3281 ft.) of Ozaphane. A 3000-meter (9842-ft.) subject would 
weigh more than 20 kilograms (44 Ib.) when made upon regular film, but only 
7 kilograms (I5 l /z Ib.) if Ozaphane were used. 

In the diazotype process, the material from which the image is to be formed may 
be invisible in the cellulosic vehicle until development, which forms a colored 
compound. The colored substance is formed when two suitable chemical sub- 
stances in the film react in an alkaline medium, which is usually provided by the 
use of ammonia gas. Premature reaction is prevented by the slight acidity 
of the medium. When light is permitted to fall upon the undeveloped film, one 
of the substances is altered in such a way as to prevent the later ammonia develop- 
ment from forming a colored compound. Thus, if Ozaphane film is printed from 
a positive, a positive image results. A direct process of this type offers some 
promise of simplifying distribution through the possibility of making duplicate 
copies from an Ozaphane exhibition print. 



* Communicated. 
** Paris, France. 
358 



APPARATUS SYMPOSIUM 359 

In addition to its use in motion picture work, Ozaphane film is expected to 
supplant the disk record as a medium for sound reproduction. For this purpose 
it is proposed to use a track 0.5 millimeter (0.0197 inch) wide, scanned at a rate of 
0.45 meter (1.476 ft.) per second. By placing nine such tracks side by side upon 
a film 9.5 millimeters (0.374 inch) wide, a 65-meter (213-ft.) length will provide 
about 22 minutes' playing time. 

THE CINELUX PROJECTOR CD-2 

A portable motion picture projector-sound reproducer has been designed for use 
with Ozaphane film. It consists of three units contained in two trunks for trans- 




FIG. 1. Front elevation of Cinelux 
projector, showing film path. 

portation. In use the projector is mounted above the amplifier, and the loud 
speaker is placed near the projection screen. The most unusual feature of the 
apparatus is the film-moving system, which utilizes friction driving rollers in- 
stead of sprockets or claws. 

The film path is arranged as follows, starting at the top of the machine (Fig. 1) : 
feed roll, upper drive roller, loop control roller, picture gate, intermittent, lower 
drive roller, sound scanning station roller, damping roller, and friction take-up. 
The film is moved by the upper drive roller at a speed just sufficient to maintain 
a loop above the picture projection gate. Regulation of the speed of film travel 
over this drive roller is effected by means of a control roller located within the 
upper loop, in such a way as to maintain a loop of sufficient size at all times. 
Velvet pads hold the film in position in the gate. At a short distance beyond the 



360 



APPARATUS SYMPOSIUM 



[J. S. M. p. E. 



gate is the lower drive roller which moves the film continuously at a constant 
rate and is the pacemaker for the rest of the machine. Between this roller and the 
gate is the intermittent of the beater roller type, which causes the pull-down by 
intermittently enlarging the loop between the gate and the lower drive roller. 
Beyond the lower drive roller is a film-driven roller over which the film runs 




FIG. 2. Diagram showing arrangement of automatic 
framing system; lever I adjusts the image upon the 
screen. 



for sound scanning. Between this point and the slippage-driven take-up is a 
vibration-absorbing roller. 

Framing is maintained automatically by means of the perforation images 
printed upon the Ozaphane film from the master positive, which has standard 
perforations. The projection aperture (Fig. 2) is cut away at one side to provide 
illumination of the area bearing the perforations. Light transmitted by this area 
passes through a prism and lens to an adjustable mask at the front of a light- 
sensitive cell. The cell operates in conjunction with an amplifying system to 
energize a magnetic brake which, in turn, causes the intermittent to be advanced 
or retarded accordingly as the framing tends to change. 



A HIGH-SPEED CAMERA 

C. T. BURKE* 



The study of mechanical transients occurring at very high speeds requires a 
camera capable of much higher picture rates than the conventional slow-motion 
camera. This problem has been attacked from several angles, but the methods 
of approach have followed along two avenues the utilization of a moving optical 
system to preserve a stationary relationship between the film and the image, and 
the utilization of a brief and brilliant flash of light to establish the exposure. 

* General Radio Co., Cambridge, Mass. 



Oct., 1935] APPARATUS SYMPOSIUM 361 

The latter approach enormously simplified the camera design problem. The 
mechanical design of a camera capable of taking pictures at rates as high as 5000 
per second presents serious problems in the design of all parts of the camera to 
give reliable operation at such rates. The elimination of any moving shutter part 
is, therefore, a distinct gain. The stroboscopic-light method of exposure makes 
possible the elimination of shutters, intermittent motions, or means for moving 
the image with the film. 

The camera is reduced to a means for carrying the film past an aperture of 
proper dimensions at the desired speed. The number of moving parts is reduced 
to film reels and drive motors. To be sure, the simplification of the camera is at 
the expense of an additional complication in the lighting system, but a considera- 
tion of all factors indicates that the stroboscope method offers the best ap- 
proach to the problem of ultra high-speed photography. 

The structure of the camera is shown in Fig. 3. The moving parts consist of 
the sprocket and two spools, the drive motors, and the commutator which con- 
trols the stroboscopic flash rate and serves to space the frames upon the film 
properly for projection (Fig. 4). 

The principal problem in the design of the drive sprocket and reels was the 
elimination of inertia and the reduction of friction, since there is some fire hazard 
'evolved in passing film through the camera at very high speeds. The matter of 
inertia is of great importance because of the necessity of bringing the camera 
quickly to speed. 

The film is pressed against the sprocket by an aluminum roller, and a metal plow 
is provided under the sprocket to prevent the film from being carried around by 
the sprocket teeth and being broken. The plow does not touch the film in normal 
operation, but immediately corrects any tendency to stick, and also prevents 
the film from winding around the sprocket in case of breakage. 

No mechanical connection is provided between the driving sprocket and the 
take-up reel. The take-up reel is driven by a series motor of sufficient torque to 
keep the film taut but not to break it. The difference required in the speed by 
the changing diameter of the reel as the film is taken up is thus automatically 
cared for by the slowing down of the motor. Extremely high acceleration 
can be provided for, since it is possible to operate the motors at over- voltage. 
The time required to run 100 feet of film through the camera is only a second or 
two, and no damage results to the motors from such momentary overloading. 
The acceleration under these conditions is remarkably rapid. Full and constant 
speed is attained within less than ten per cent of the film length. 

A standard lens mount focuses the image behind the aperture, which is fitted 
with a F-edged gate. Focusing is facilitated by a diametric opening in the 
sprocket and an eye-piece in the back of the camera. 

It will be observed that distortion in the image will result in consequence of the 
curvature of the sprocket. This distortion can be minimized by the use of a suf- 
ficiently large sprocket diameter. In the camera described, the sprocket diameter 
is 4.75 inches, carrying twenty standard 35-mm. frames upon its periphery. The 
center of the picture is 0.0295 inch closer to the lens plane than the extreme up- 
per and lower edges of the frame. This difference does not cause serious distor- 
tion with the subjects and lens systems normally used. When using 16-mm. film 
the distortion is, of course, proportionately reduced. 



362 



APPARATUS SYMPOSIUM 



[J. S. M. p. E. 




. 

I* 




Oct., 1935] APPARATUS SYMPOSIUM 363 

Since the film moves through the camera continuously without shutters, the 
distribution of exposures upon the film is determined by the rate of flash of the 
stroboscope light in relation to the film speed. If the film is to be projected, the 
images must be spaced properly along the film. In order to accomplish this a 
commutator is mounted upon the sprocket shaft to provide electrical impulses 
for setting off the stroboscopic flash at equal intervals. A film thus exposed can 
be projected in standard projection equipment. The commutator presents a 
rather serious mechanical problem. The segments must be located with extreme 
precision. Any irregularity in their placement will appear as flicker in the pro- 
jected picture. The film speed is so great that a minute angular displacement of 
the commutator segment will result in rather serious flicker difficulty. 

This camera has important applications in the study of high-speed mechanisms 
and various mechanical transients. Picture rates (using 16-mm. film) as high as 
5000 per second have been attained. These films may be reprojected, giving an 
extraordinary slowing down of the motion observed. 



THE WALL MOTION PICTURE CAMERA 

H. GRIFFIN* 

The Wall camera (Fig. 5) was originally developed in 1926 and 1927 for newsreel 
work and has recently been redesigned for use either in the studio or in the field. 
The camera can be equipped with either light-valve, flashing lamp, or galvanom- 
eter. It is driven by a direct-drive, 12-volt, d-c. motor of approximately Vso hp. 
The motor can be of the interlocking type, if desired, to run with a 12-volt motor 
of the same type upon a sound recorder. The camera can also be furnished with 
a synchronous motor for synchronizing with a sound recorder, if desired. Each 
motor has but one bearing fitted at the outer end of the armature shaft. The 
inner end of the armature shaft is direct coupled to the shutter shaft, providing a 
direct form of drive. The outer end of the shutter shaft constitutes the inner 
bearing for the armature shaft. 

The sprocket is run by the shutter shaft through a worm and gear of the hunt- 
ing-tooth type and of the proper ratio to use a 31-tooth sprocket. The reason for 
this is that if the film, which is perforated four holes at a time, should be unevenly 
spaced, the hunting tooth of the sprocket would minimize whatever error oc- 
curred. The teeth of the sprocket are made narrower on the sound side than 
standard, so as to eliminate interference with the sound-track. 

The film trucks are made to give a l /^'mch opening for threading, and are pro- 
vided with safety pins at the outer ends so that the door can not be closed unless 
they are in position. 

t;he intermittent is driven from the sprocket shaft through high-helix-angle 
gears to minimize any inequality of motion imparted by the gearing. This inter- 
mittent is of the claw and locking-pin type, using three cams and running in 



"International Projector Corp., New York, N. Y, 



364 



APPARATUS SYMPOSIUM 



[J. S. M. p. E. 





Oct., 1935] APPARATUS SYMPOSIUM 365 

closely fitted ways. The aperture plate is made of stainless steel, hardened, 
ground, and lapped to a mirror finish. A spring-actuated pressure pad with 
three bakelite rolls is the only pressure contact exerted upon the film in its passage 
through the intermittent. The aperture and back plate have a 0.002-inch 
clearance over the film. The pull-down cam of the intermittent has a dwell of 
approximately 202 degrees, which makes the shutter opening approximately 190 
degrees. The particular method of installing the intermittent in the camera 
provides a further dwell of 33 degrees, making a total shutter opening of 225 
degrees. 

The shutter is provided with an adjustable blade operated through right- and 
left-hand spirals upon the shutter shaft. These spirals are so made that, if de- 
sired, they will produce a complete fade-out with a 180-degree shutter. In the 
case of a 225-degree shutter opening, enough of the spiral is used to make the 
shutter adjustable from 90 to 225 degrees. The intermittent can be opened for 
threading, and by the turn of a wing-screw can be removed from the camera. It 
is provided with a case that covers the cams and has a removable opening in the 
top. A special grease is used which will adhere to the cams for several months 
without renewal. 

The turret is provided with openings for four lenses and locks in a hardened 
and ground F-shaped slot upon the periphery of the turret. The locking device 
is provided with a coiled spring to eliminate the shock of indexing. It is also 
provided with a knurled adjusting nut to shift the lenses about the axes of the 
turret for angle shots. The lenses are attached to the turret with a bayonet lock, 
and are so made that the barrel of the lens does not turn over when focusing the 
lens. The graduation of the lenses can be read from the operating position. 
Unless otherwise specified, Bausch & Lomb lenses of the following sizes are 
used: 40- or 50-mm., //2.3; 75-mm. and 100-mm.,//2.3; and 152-tnm., //2.7. 
The recently developed 25-mm.,//2.3 Bausch & Lomb lens can also be supplied. 

The focusing device is of the shift-over type ; that is, the camera proper slides 
over upon the base so that the focusing tube is directly back of the photo- 
graphic lens. When the operator wishes to follow an object while photograph- 
ing it, he does so by using the same focusing device with a wide-angle lens mounted 
directly ahead of the focusing tube and having a glass with the various aperture 
sizes etched thereon to give the exact position of the photographed object. The 
camera is provided with a footage counter and a tachometer for registering the 
speed of the camera. It is also provided with a handle suitably placed so that the 
weights of the camera, motor, and lenses balance. All parts of the camera except 
the intermittent are oiled by three oilers at the back of the camera. The weight 
of the camera, motor, and lenses is forty-four pounds. When the camera is fur- 
nished with the flashing lamp, it is supplied with a lamp holder, quartz, and quartz 
shoe accurately positioned with respect to the film, the sound being registered 
directly upon the sprocket. When furnished with a light-valve or galvanometer, 
the units are so made as to be interchangeable in position with the flashing lamp. 

The magazine is equipped with ball bearings throughout, has a removable light- 
trap to facilitate cleaning, and is operated manually by a knob upon the door of 
the camera. The spools are collapsible and are easily removable from the roll of 
film. The weight of the camera, tripod, loaded magazines, motor, and lenses is 
considerably less than eighty pounds. 



366 APPARATUS SYMPOSIUM 

TRIPOD 

The tripod is made with a flat top when used with the standard camera, and with 
a dove-tail slide when used with the special camera. When made with the flat 
top, the hold-down screw is in the center of the top and is operated by means of a 
knurled knob from the side of the head. The tilt is approximately 40 degrees 
either side of the center. The head is operated in the usual manner by pressure 
exerted upon the handle. 

The action of the tripod differs from that of most other devices of this kind in- 
asmuch as the driving member for both the pan and the tilt is a worm-gear driving 
a worm upon which the damping friction is applied, the principle being that to 
obtain a steady movement the friction must be applied close to the center of 
the worm, and while the fine adjustment necessary when friction is applied 
over a large surface is eliminated. 



FALL, 1935, CONVENTION 

WASHINGTON, D. C. 

WARDMAN PARK HOTEL 

OCTOBER 21-24, INCLUSIVE 

Headquarters 

The headquarters of the Convention will be the Wardman Park Hotel, where 
excellent accommodations and Convention facilities are assured. Registration 
will begin at 9 A.M., Monday, October 21st. A special suite will be provided for 
the ladies. Rates for S. M. P. E. delegates, European plan, will be as follows: 

One person, room and bath $3.00 

E Two persons, double bed and bath 5. 00 

Two persons, twin beds and bath 5.00 

[ Rates for connecting parlors 5 . 00 

A modern fire-proof garage is located on the Hotel property, and a special 75 
cents per day rate has been arranged for. 

Technical Sessions 

An attractive program of technical papers and presentations is being arranged 
by the Papers Committee. Sessions will be held in the Little Theater of the Hotel, 
off the west lobby, as follows: Monday to Thursday mornings, inclusive; and 
Monday, Tuesday, and Thursday afternoons. 

Film Programs 

Exhibitions of newly released motion picture features and short subjects will be 
held in the Little Theater on Monday and Tuesday evenings. Passes to various 
motion picture theaters in Washington will be available to the members register- 
ing for the duration of the convention. 

Apparatus Exhibit 

An exhibit of newly developed motion picture apparatus will be held in the east 
lobby of the Hotel, to which all manufacturers of equipment are invited to con- 
tribute. The apparatus to be exhibited must either be new or contain new fea- 
tures of interest from a technical point of view. Information concerning the 
exhibit and reservations for space should be made in writing to the Chairman of 
the Exhibits Committee, Mr. O. F. Neu, addressed to the General Office of the 
Society at the Hotel Pennsylvania, New York, N, Y , No charge will be made 
for space. 

367 



368 FALL, 1935, CONVENTION [j. s. M. P. E. 

Informal Get-Together Luncheon 

The usual luncheon will be held at noon on October 21st in the Continental Room 
of the Hotel. An address of welcome will be delivered by the Honorable Sol 
Bloom, member of Congress from New York. Other speakers will be announced 
later. 

Semi- Annual Banquet 

The semi-annual banquet of the Society will be held in the Continental Room 
of the Hotel on Wednesday October 23rd at 7 : 30 P.M. Addresses will be delivered 
by eminent members of the industry followed by dancing and entertainment. 
The presentation of the scroll of honorary membership to Thomas Armat, of 
Washington, D. C., awarded last May at Hollywood, will be made, and, in addi- 
tion, the recipients of the Journal Award and the Progress Medal of the Society 
will be announced and the presentations made. 

Points of Interest 

To list all the points of interest in and about Washington would require too 
much space, but among them may be mentioned the various governmental 
buildings, such as the Capitol, the White House, Library of Congress, Depart- 
ment of Commerce, U. S. Treasury, U. S. Bureau of Standards, Department of 
Justice, Archives Building; and other institutions such as the National Academy 
of Sciences, the Smithsonian Institution, George Washington University, Wash- 
ington Cathedral, Georgetown University, etc. In addition may be included the 
Lincoln Memorial, the Washington Monument, Rock Creek Park, The Francis 
Scott Key Memorial Bridge, Arlington Memorial Bridge, the Potomac River 
and Tidal Basin. Mt. Vernon, birthplace of Washington, is but a short distance 
away and many other side trips may be made conveniently via the many high- 
ways radiating from Washington. 

Recreation 

The Wardman Park Hotel management is arranging for golfing privileges for 
S. M. P. E. delegates at several courses in the neighborhood. Regulation tennis 
courts are located upon the Hotel property, and riding stables are within a short 
distance of the Hotel. Trips may be arranged to the many points of interest in 
and about Washington. 

PROGRAM 
Monday, Oct. 21st 

9 : 30 A.M. Registration 
10:00 A.M. Society business 

Technical papers program 
12:30 P.M. Informal get-together luncheon 

2:00 P.M. Technical papers program 

8:00 P.M. Exhibition of newly released motion pictures 



Oct., 1935] FALL, 1935, COISTVENTION 369 

Tuesday, Oct. 22nd 

10: 00 A.M. Technical papers program 
2 : 00 P.M. Technical papers program 
8:00 P.M. Exhibition of newly released motion pictures 

Wednesday, Oct. 23rd 

10:00 A.M. Technical papers program 

12:30 P.M. Free afternoon, for recreation or special trips and visits 
7: 30 P.M. Semi-annual banquet 

Thursday, Oct. 24th 

10: 00 A.M. Technical papers program 
2:00 P.M. Technical papers program 
6:00 P.M. Adjournment of the Convention 



SOCIETY ANNOUNCEMENTS 

BOARD OF GOVERNORS 

At a meeting held at the Hotel Pennsylvania, New York, N. Y., September 
13th, further work was done in drafting Administrative Practices, a compendium 
of all the Board actions in effect and not specifically prescribed in the Constitution 
and By-Laws. Administrative Practices will be kept up to date from year to year, 
and will thus constitute a complete code of currently operative policies for the 
guidance of each successive Board. Included are the regulations pertaining to 
the various Awards of the Society, Honorary Membership, Standardization pro- 
cedure, etc. 

OFFICERS FOR 1936 

Ballots for voting for officers of the Society for 1936 were mailed recently to the 
voting membership. The nominations were as follows : 

H. G. TASKER, President J. H. KURLANDBR, Secretary 

E. HUSE, Executive V.-P. T. E. SHEA, Treasurer 

L. A. JONES, Engineering V.-P. A. S. DICKINSON, Governor 

O. M. GLUNT, Financial V.-P. H. GRIFFIN, Governor 

A. C. HARDY, Governor 

The President, Executive Vice -President, Secretary, and Treasurer hold office 
for one year; the other Vice-Presidents and the three Governors hold office for 
two years. 

Present members of the Board whose terms do not expire until December 31, 
1936, are: 

A. N. GOLDSMITH, Past-President W. C. KUNZMANN, Convention V.-P. 

J. I. CRABTREE, Editorial V.-P. M. C. BATSEL, Governor 

S. K. WOLF, Governor 

Results of the election will be announced at the Fall Convention at Washing- 
ton, D. C., October 21st, and the successful candidates will assume office on 
January 1, 1936. 

SECTIONAL COMMITTEE ON 
MOTION PICTURES UNDER THE ASA 

As announced previously, no definite technologic decisions were reached at the 
Paris Congress, although an agreement was achieved as to the manner in which 
16-mm. standardization was to be handled. The Deutschen Normenausschuss, 
a member of the International Standards Association (ISA) was appointed ISA 
Secretariat for the project, and questionnaires have been mailed to the nineteen 
national standardizing bodies, in as many countries, who are members of 
the ISA. 
370 



SOCIETY ANNOUNCEMENTS 371 

The items of the Questionnaire are as follows : 

(1) Have you a Committee for the standardization of 16-mm. sound-film? 
(Yes or No.) 

(2) If not, are you going to institute such a Committee? (Yes or No.) 

(3) Which Associations of your country deal with the question? 

(4) Which of these Associations are collaborating with you? 

(5) Do you agree with Recommendation Bl, as to the distance between the 
sound and the picture? ( Yes or No.) If not, please state the reasons. 

(Recommendation Bl: The distance between the sound and the corresponding 
picture should be 26 pictures. The existing standards of 25 (SMPE) and 27 (DIN) 
pictures, respectively, should be changed to the standard of 26 as soon as a revision 
becomes possible.) 

(6) Do you agree with Recommendation B2 as to the position of the emulsion 
in the projector? ( Yes or No.) If not, please state the reasons. 

(Recommendation B2: (a) Film obtained by reversal -emulsion toward the lens, 
(b) 16- Mm. positive film obtained by contact printing from a 16-mm. negative 
emulsion toward the source of light, (c) 16- Mm. positive film obtained by optical 
reduction from 35 -mm. films may be placed with the emulsion side either to the lens 
or to the light-source, (d) In order to permit projection of 16-mm. sound-films, 
whether obtained by the reversal process, optical reduction, contact printing, or color 
processes, it is recommended that the sound-head be provided with a device allowing 
refocusing of the sound objective, according to whether the emulsion is on one side or 
the other. 

(7) Which position of the sound-track do you propose (i. e. t SMPE standard 
left; or DIN-ICE standard right)? 

(8) Which Associations of your country have agreed with your proposal? 

(9) Are you prepared to accept the opposite position of the sound-track, if 
this represents the votes of a majority of nations? ( Yes or No.) If not, please 
state the reasons. 

(10) What is the amount of capital invested in your country in (a) 16-mm. 
sound-film apparatus sold and at present in production; (b) 16-mm. sound-films 
sold and at present in production -e. g., in archives, etc.; (c) development work 
and manufacturers' equipment tools, etc. 

This questionnaire will be acted upon shortly by the Sectional Committee, 
and then forwarded to the American Standards Association for transmittal to the 
ISA and the Secretariat. 

SOCIETY SUPPLIES 

Reprints of Standards of the SMPE and Recommended Practice may be obtained 
from the General Office of the Society at the price of twenty-five cents each. 

Copies of Aims and Accomplishments, an index of the Transactions from October, 
1916, to June, 1930, containing summaries of all the articles, and author and 
classified indexes, may be obtained from the General Office at the price of one 
dollar each. Only a limited number of copies remains. 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 



372 SOCIETY ANNOUNCEMENTS 

JOURNAL, may be obtained from the General Office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



The Society regrets to announce the death of one of its members : 

ABDULLA FAZALBHOY 

July 15, 1935 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV NOVEMBER, 1935 Number 5 



CONTENTS 

Page 

The Photographic Effectiveness of Carbon Arc Studio Light- 
Sources F. T. BOWDITCH AND A. C. DOWNES 375 

The Radiant Energy Delivered on Motion Picture Sets from 

Carbon Arc Studio Light-Sources 

F. T. BOWDITCH AND A. C. DOWNES 383 

Modern Instruments for Acoustical Studies. . . .E. C. WENTE 389 

Flutter in Sound Records 

T. E. SHEA, W. A. MACNAIR, AND V. SUBRIZI 403 

A Portable Flutter-Measuring Instrument. . . .R. R. SCOVILLE 416 
Lighting for Technicolor Motion Pictures . . . . C. W. H ANDLEY 423 
Report of the Studio Lighting Committee 432 

The Use of Films and Motion Picture Equipment in Schools .... 

MARION EVANS 443 

New Apparatus a New High-Fidelity Sound Head 

F. J. LOOMIS AND E. W. REYNOLDS 449 

Society Announcements 461 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE. Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York. N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDKBY K. WOLF, 250 W. 57th St., New York, N. Y. 



THE PHOTOGRAPHIC EFFECTIVENESS OF CARBON ARC 
STUDIO LIGHT-SOURCES* 



F. T. BOWDITCH AND A. C. DOWNES** 



Summary. The work done in connection with the photographic effectiveness oj 
National motion picture studio carbons has been extended to include the Sun Arc car- 
bons and 1 / 2 x 12-inch Rotary Spot carbons. 

Energy distribution curves of the two latter types of carbon at various currents are 
given, and the beneficial effects of increased current upon the photographic effectiveness 
of these two types of light-sources are emphasized. The desirability of the higher 
currents in using these sources is very clearly demonstrated. Included are the effects 
of the use of a gelatin filter frequently used with the Sun Arcs and Rotary Spot carbons. 

The value of a light-source for illuminating a photographic subject 
depends upon its spectral distribution; that is, upon the proportions 
of violet, blue, green, yellow, orange, and red light in the radiant 
energy emitted by the source; and upon the spectral or color-sensi- 
tivity of the photographic film or plate upon which the picture is re- 
corded. Previous communications 1 ' 2 have given the spectral distri- 
bution and photographic effect of certain carbon arc light-sources 
which have been or are being used in motion picture and other photo- 
graphic studios for various types of photography and photo-engraving. 
Additional information on other light-sources used primarily in the 
motion picture studios is given here. 

The previous papers dealing with carbon arc light-sources for photog- 
raphy have dealt with flame arcs which may be used with either 
direct or alternating current. The additional light-sources described 
in this paper are of the high-intensity type. The distinguishing fea- 
tures of these two distinctly different types of arcs have been de- 
scribed previously in this JOURNAL 3 . The high-intensity light-sources 
used in the studios are the so-called sun arcs and rotary spots, which 
can be operated only on direct current. The spectral distributions of 
these light-sources are somewhat similar, but differ quite materially 
in their photographic effectiveness. 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Research Laboratories, National Carbon Co., Cleveland, Ohio. 

375 



376 



F. T. BOWDITCH AND A. C. DOWNES [J. S. M. P. E. 



Fig. 1 shows the spectral energy distribution of 16-mm. high- 
intensity sun arc carbons operated at 130, 140, and 150 amperes. 
From these curves it will be readily appreciated that the form of the 
energy distribution of the high-intensity sun arc does not vary ma- 




VOLET BLUE 



GREEN YELLOW (jiRANGE. 



I404MPS 75V 
-I30AMPS 76V 



5000 6000 

WAVE LENGTH IN ANGSTROM UNITS 



FIG 1. Spectral energy distribution of motion picture sun 
arcs; positive crater radiation only: 16-mm. H. I. positive, 
11-mm. H. I. negative. One square = 250 microwatts per sq. 
cm. at 10 feet. 



terially with the current, but the differences that do exist are sufficient 
to cause the photographic effectiveness, as will be shown later, to be 
affected to some degree by the current. 

Fig. 2 shows the spectral energy distribution curves of J /2 by 12- 
inch rotary spot carbons at 70, 80, 90, and 110 amperes. These 




5000 6000 

WAVELENGTH IN ANGSTROM UNITS 



FIG. 2. Spectral energy distribution of Ashcraft rotary 
spot arc; positive crater radiation only: Y 2 by 12 rotary 
spot positive, */ CC Orotip negative. One square = 125 
microwatts per sq. cm. at 10 feet. 



curves also have essentially the same form, but here again the photo- 
graphic effectiveness is materially affected by the current. 

In Fig. 3 are reproduced the spectral energy distribution curves of 
8-mm. motion picture studio carbons given in a previous paper. 2 

Table I lists the actual amounts of energy in sunlight at three 



Nov., 1935] 



EFFECTIVENESS OF LIGHT-SOURCES 



377 



ooo 

1 ct I 



00 o o o o 
O^ ^ CO lO 00 CO 

oT oo" 10" ^" co" c^ r-T 



o cn> o 
2i 



T-H OiOO OCT>iOI> 

TJH 100606 oi^cooo 

CO CO CO CO CO "^ ^f CO 



So o ooo oooo 

O (M OCO^O OiO5COCO 

T4 QQ QQ l^) ^ ^) tO ^* C^ C^J 

00"l>" co" co"co"c<f r-T i-T I-H" rl? 



OOCOt>Oi 



000 
r*- O^ iO 

t>r 10" co" 

CO CO CO 



%%% %%%% 

00 oooo 
O5 CO lO lO Tt< TJH 



1> lO CO 

cq d 06 

ooo 

rr 1 W 3? 



oocdcd 

<M<MCO 



II 

p 1 "' 



o ooo oooco 



II 



OCOO1> 

lOiOiOCO 



I 



m <M 

> ^ 

Sx 

o *> 



05 



378 



F. T. BOWDITCH AND A. C. DOWNES [ J. S. M. P. E. 



seasons of the year, and of the three types of arc lamps used in motion 
picture studios as just described. The table gives the total energy for 
each of these sources between 3400 A, the cut-off of the glass camera 









_^ 


-^ 





^-^__ 
















J 


" 


^^ 










=: 


^ 














3000 


4000 5000 6000 /UOO 



FIG. 3. Spectral energy distribution, 8-mm. National CC 
motion picture studio carbons; 35 and 40 amps., 37.5 volts 
d-c. One square = 250 microwatts per sq. cm. at 1 meter. 

lens in the ultraviolet, and 50,000 A, the limit of sunlight and of the 
transmission of glass in the infrared portion of the spectrum. In 
addition to the total radiant energy, the table shows the actual 
amounts and percentages in three spectral bands, viz., the photo- 
graphically effective, from 3400 to 7000A (including visible light from 



i^ 



CREfrl 



- SUNLIGHT 



~S5S5 6<Sx5 

LENGTH IN ANGSTROM UNITS 



-130A S W ARC+FirER '53 \^RY 
LIGHT STRAW 



-0-150 V SUN AH 



\ 

RED 



INTRA-RdP 



FIG. 4. Relative photographic effects of 16-mm. H. I. carbon 
sun arcs on Supersensitive panchromatic film. 

4000 to 7000 A) ; the penetrating infrared from 7000 to 14,000 A, 
and the long-wave infrared from 14,000 to 50,000 A. 

The photographically effective band is the only one of any value for 
ordinary photography, and the other two bands are usually referred 
to as heat. The most efficient photographic light-source is, therefore, 



Nov., 1935] 



EFFECTIVENESS OF LIGHT-SOURCES 



379 



the one having the highest percentage of its energy in the photo- 
graphically effective band from 3400 to 7000A. 

Curves of photographic effect, taking into account the spectral 



ULTRA-VOLET VOLET BLUE 



- - IIOA 63V*- FILTER '53 VERY LIGHT STRAW 
70A 50V ^FILTER ' 53 VERY LIGHT STRAW 
-- IK) A, 61V 
-e- 90A 55V 



YELLOW pRANGE 



SUNLIGHT 



70 A 50V 



5000 6000 

WAVE LENGTH IN ANGSTROM UNITS 



FIG. 5. Relative photographic effect of l /% by 12 rotary spot 
carbons on Supersensitive panchromatic film. 

energy distribution of the light-source, the spectral sensitivity of 
Eastman supersensitive panchromatic film, and the transmission of 
the camera lenses, have been calculated by a method described by 
Jones. 4 The results of these calculations in the case of the high- 





















. 




*^-~~~~~ 








TRANSMISSION 
2 


/ 


\ 




7 














"TV 




















ER CENT 












































3000 4000 5000 600O 7OOO 



WAVE LENGTH IN ANGSTROM UNITS 



FIG. 6. Light transmission of gelatin filter No. 53 Very Light 
Straw (Brigham Gelatin Co.). 



intensity carbons used in the sun arcs are shown in Fig. 4. Included 
in this illustration is the photographic effect of sunlight, for compari- 
son with the carbon arcs. The curves show that the photographic 
effect of the sun arc at 130 amperes is not quite as great as at 150 
amperes. The effect of operating the lamp at 140 amperes, however, 



380 



F. T. BOWDITCH AND A. C. DOWNES [J. S. M. P. E. 



is essentially the same as that of operating at 150 amperes, except 
that the total amount of light is greater at the higher current. 

Since these sun arcs are frequently used with niters, particularly in 
color motion picture photography, there are included in Fig. 4 the 
relative photographic effects of the 130-ampere arc plus the Brigham 
Gelatin Company's gelatin filter No. 53, Very Light Straw. The 
effect of this filter is to reduce the violet and the blue, and to accen- 
tuate the orange and the red effectiveness of the light-source, as a 
comparison of the curves will show. 

In Fig. 5 are shown the similar curves of photographic effectiveness 



ORWNATES REPRESENT THE PRODUCT OF LEGEND 
TRANSMISSION OF GLASS LENS. SUNLIGHT 
RELATIVE SENSITIVITY OF FILM AND 8MM NATIONAL CC MP STUDIO C 
SPECTRAL ENERGY DISTRIBUTION OF LIGHT SOURCE 40 AMPS 37 5 VOLTS 
VALUES ADJUSTED TO GIVE MAXIMUM' 100 






ARBONS 


ULTRA-VIOLET VIOLET BLUE GREEN YELLOW ORANGE RED 


100 








' 




\ 




















- x 




















j 


,' 


\ 










, 


-- 


, 




/ 




\ 










fQ U 














\ 




















\ 
\ 










*S 7 








1 








^ 








/ 


x 




/ 


"N 


\ 










1 
M J 

,00 
J 

?0 u 






/ 


; 










\ 


^._ 


/ 












\ 














/ 






























\\ 










































\ 












I 






























v 








5 






/ 


































\ 






L 






































\ 







WAVE LENGTH IN ANGTROM UNITS 

FIG. 7. Photographic effect in relation to wavelength, for 
Supersensitive panchromatic film. 



for rotary spot carbons at 70, 90, and 110 amperes. The photographic 
effects at 70 and 110 amperes are shown both with and without the 
filter No. 53, Very Light Straw. The curve for 90 amperes is without 
the filter. The curves show that the effect of increasing the current 
upon the photographic effect of the 1 / z by 12 rotary spot carbons is 
considerably greater than in the case of the sun arc carbons. 

The work at 110 amperes was done only to emphasize the result of 
increasing the current upon the photographic effectiveness, and 
should not be understood as a recommendation to burn 1 /z by 12 
rotary spot carbons at this high current. It is hoped that these figures 
will encourage the studio technicians to operate their various carbon 
arc lamps at the highest practicable current in each case, which is 



Nov., 1935] 



EFFECTIVENESS OF LIGHT-SOURCES 



381 



about 150 amperes for 16-mm. carbons in the sun arcs, and 90 to 95 
amperes for the 1 / 2 by 12 rotary spot carbons. 

Particular attention should be given to the fact that the photo- 
graphic effect of rotary spot carbons at 110 amperes with no filter is 
better than that of the same carbons at 70 amperes with the No. 53 
filter. 

Fig. 6 shows the percentage transmission of the Brigham Gelatin 
Company's gelatin filter No. 53 Very Light Straw. While such a filter 
results in at least partial correction of the greater photographic effect 
of the blue and violet, or short-wave energy, it should be remembered 
that it actually reduces the quantity of light available. Fig. 7 is a 




FIG. 8. Relative quantities of penetrating and long-wave 
infrared radiation, for equal quantities of photographically 
effective radiation. 

A = 3400-7000 A photographically effective 

B = 7000-14,000 A penetrating infrared 

C = 14,000-50,000 A long-wave infrared 

reproduction of the curve of photographic effect of the 8-mm. National 
motion picture studio carbons given in the article previously cited. 2 

The spectral energy distribution curves that have been given cover 
only the near ultraviolet and the visible parts of the spectrum, since 
these include the only wavelengths that are effective in reproducing 
the photographic image upon supersensitive panchromatic films or 
plates. It should not be forgotten, however, that all light-sources 
have the greater proportion of their radiant energy in the penetrating 
and long-wave infrared regions. These portions of the spectrum differ 
in their characteristics, particularly by their ability to pass through 
water or human flesh, but are usually classed together as heat. 

Fig. 8 shows the relative amounts of penetrating and long-wave 
infrared radiations emitted for equal amounts of photographically 



382 F. T. BOWDITCH AND A. C. DOWNES 

effective radiations, including the visible spectrum plus a small quan- 
tity of near-ultraviolet radiation. In this illustration, for each of the 
light-sources, the quantity of photographically effective radiation 
from 3400 to 7000 A has been fixed at 100, and the quantities of the 
penetrating and long-wave infrared bands are plotted as percentages 
of the photographically effective region. 

In the cases of the 16-mm. high-intensity carbons used in the sun 
arcs, and the V2-mch carbons used in the rotary spots, the effect of in- 
creased current is very noticeable in decreasing the relative amounts 
of penetrating and long-wave infrared radiations. It is therefore ap- 
parent, from the data presented in this paper, that much is to be 
gained by running the high-intensity arcs at the highest possible cur- 
rent within the rating of the carbons. Not only is the photographic 
efficiency of the light much improved, but the infrared radiation per 
unit of photographic light, which has no useful effect, is materially 
reduced. 

REFERENCES 

1 JOY, D. B., AND DOWNES, A. C.: "Characteristic of Flame Arcs for Studio 
Lighting," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 34, p. 502. 

2 JOY, D. B., BOWDITCH, F. T., AND DOWNES, A. C.: "A New White Flame 
Carbon for Photographic Light," /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), 
No. 1, p. 58. 

*JoY, D. B., AND DOWNES, A. C.: "A New Alternating-Current Projection 
Arc," J. Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 116. 

4 JONES, L. A.: "Use of Artificial Illuminants in Motion Picture Studios," 
Trans. Soc. Mot. Pict. Eng., V (1921), No. 13, p. 74. 



THE RADIANT ENERGY DELIVERED ON MOTION PICTURE 
SETS FROM CARBON ARC STUDIO LIGHT-SOURCES* 



F. T. BOWDITCH AND A. C. DOWNES** 

Summary. The spectral energy distributions are given of Sunlight, National 
motion picture studio carbons, Sun Arcs, and Rotary Spot carbons under the condi- 
tions found in the studios. From the spectral energy distributions, the amounts of 
actual energy received by an actor from one of these light-sources at various distances, 
and the numbers of carbon arc lighting units required to equal the intensity of sunlight 
have been calculated, 

A chart is included which shows the relative amounts of photographically effective, 
penetrating infrared and long-wave infrared radiations emiHed by the various arc 
units. 

The substitution of artificial for natural lighting in the production 
of most of the scenes used in modern motion pictures has often raised 
the question as to how the artificial light-sources used in the studios 
compare with natural sunlight as to quality (spectral composition), 
particularly in respect to the division between visible light or photo- 
graphically effective radiant energy and the penetrating and long- 
wave infrared, commonly classified together as "heat." This article 
discusses these questions for the various types of carbon arcs used for 
photographic purposes. 

Motion picture studios use three types of carbon arc lighting units, 
the "sun arcs," burning 16-mm., high-intensity carbons at 130 to 150 
amperes; the "rotary spots," burning 1 / z by 12 rotary spot carbons 
at 70 to 80 amperes; and the "side arcs," burning 8-mm. National 
M. P. studio carbons at 35 to 40 amperes. "Domes" and "scoops" 
are, for the purposes of this article, included in the "side arc" classifi- 
cation. The "sun arcs" and "rotary spots" are "high-intensity" arcs, 
and the "side arcs" are ordinary "flaming arcs." The distinguishing 
features of these two types of arcs have been described in previous 
communications. 1 The motion picture studios use only direct cur- 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Research Laboratory, National Carbon Co., Cleveland, Ohio. 

383 



384 



F. T. BOWDITCH AND A. C. DOWNES [ J. S. M. P. E. 



rent, and while the high-intensity arcs of the type used in the "sun 
arcs" and "rotary spots" operate only on direct current, the "flaming 
arcs," such as are used in the "side arcs," operate satisfactorily on 
alternating current with no great loss in output of radiant energy. 

The spectral characteristics of these different types of arcs can best 
be shown by means of the curves. Fig. 1 shows (a) the spectral dis- 
tribution of the radiant energy in natural June sunlight, essentially 
like that of Southern California; (b) the radiant energy from a 150- 
ampere, high-intensity carbon sun arc at a distance of 10 feet; (c) the 
radiant energy of a 70-ampere carbon arc rotary spot at 10 feet; and 
(d) that of a National motion picture studio carbon at 3.28 feet (one 



150 30 












SPECTRAL ENERGY DISTRIBUTION CURVES 


21 

mr> ?n 






rSUN 


LIGHT 

~-\ 


v 


< 


5UNLIGH1 

SUN ARC 
TARYSPC 

SIDE ARC 

t\ 


ONE 
> TS ONE 
S ONE 

s. 


D IOC 

D=s 

D_IOC 


SS8T TTSPER 

3 MICROWATTS PER 
3. CM. AT IO FT 

) MICROWATTS PER 
3. CM. AT 328 FT 




/ 






x 


RC 

>^ < 


5 bUNLIGHT 

5 5 




/ 




A NAT. M 
. AT 


= STUDIC 
328 FT. 


^s 

CARBON 




/ 





^s 


^-~ _ 

"-~- m 






-60 AMF 


K 

SUN AF 
T IO FT. 


k 






5 

3C) 


// 










---v-> 




^ 




^ 


n f 


V. 


r 70 AMR ROTARY 
i i!i 


SPOT AT IO FT 

-_-+--- 












OO 4000 50CX) 6000 7000 8000 9000 KDOOO 11000 12000 13000 I4O< 

FIG. 1. Spectral characteristics of various types of arcs. 



meter). A spectral range of about 3400 A, the ultraviolet cut-off of 
glass, to 14,000 A, the limit of penetrating infrared radiant energy, is 
covered. Examination of these curves shows that the quality of the 
radiant energy from the three types of carbon arcs is very similar to 
that of natural sunlight. 

At other currents, the spectral characteristics of the arcs differ from 
those shown in Fig. 1 only in respect to the level of the curve, and do 
not differ in form, excepting that with increasing current there is a 
tendency for the percentage of visible or photographically effective 
radiation to increase at the expense of the penetrating and long- wave 
infrared portions of the spectrum. 



Nov., 1935] 



ENERGY FROM LIGHT-SOURCES 



385 



These curves of spectral energy distribution can be used to calcu- 
late the total of the visible light or photographically effective radiant 
energy, and that of the penetrating infrared radiations. To obtain 
the complete story, however, it is necessary to add to the data ob- 
tained from the curves the non-penetrating or long-wave infrared 
radiation beyond 14,000 A, which is also emitted by sources of radiant 
energy and manifested as heat. 

In Fig. 2 are shown the amounts of radiant energy delivered upon a 
motion picture set by the carbon arc light-sources of Fig. 1. This 



I-' 6 
L. 

SJ J 

CL 

B 
in 

I 3 

I 2 

I 

FIG. 



5551 SUNLIGHT 
MAY-.1UNF 



853 



RADIANT ENERGY RECEIVED ON AN AREA 
OF ONE SQ. FT. 

FROM 
- STUDIO CARBON ARC LIGHT SOURCES 



A -3400-7000 A PHOTOGRAPHICALLY EFFECTIVE 
B -7000- 14000 A PENETRATING INFRARED 
" C -14000- 50000 A LONG WAVE INFRARED 




ABC ABC ABC ABC ABC A 

2. Radiant energy delivered upon sets by the carbon arc light-sources of 

Fig. 1. 



energy is shown in three divisions, as follows: (A) the photographi- 
cally effective, from 3400 to 7000 A (including visible light from 4000 
to 7000 A); (B) the penetrating infrared, from 7000 to 14,000 A; and 
(C) the long-wave, non-penetrating infrared, from 14,000 to 50,000 A. 
These divisions are chosen because each has a characteristic behavior 
in its effect upon studio photographic materials and upon human 
sensibilities. The amounts of energy in these bands, which is the 
radiant energy upon the motion picture set, are given in gram calories 
per square foot per second. 

The effect of band A, the photographically effective and visible 
light, upon photographic emulsions and upon the human eye is 
familiar. The effects of the other two spectral bands are perhaps not 



386 



F. T. BOWDITCH AND A. C. DOWNES [J. S. M. P. E. 



so well known, and it should be noted that the demarcation between 
them is not sharp, as the limits selected here would indicate. The 
penetrating infrared, band B, which is defined as extending from 7000 
to 14,000 A, possesses the property of passing through water and also 
human flesh. The property of penetrating human flesh is not, how- 
ever, restricted to the non- visible infrared, but extends some distance 
into the red portion of the visible spectrum, as is easily demonstrated 
by holding a lighted incandescent lamp in the closed hand. 

TABLE I 

Lamp Units Required to Equal Intensity of Sunlight 



Source 


Amperes 


To Equal Photo- 
graphically Effective 
Energy of Sunlight 


Total 
Energy of Sunlight 


10-Ft. 20-Ft. 
Spot Spot 


10-Ft. 20-Ft. 
Spot Spot 


Sun Arcs 


150 
140 
,130 


3.5 14 
4 16 
4.6 18+ 


3.0 12 

3.25 13 
3.6 14+ 


Rotary Spots 


110 
90 
80 
70 


3-Ft. 30-Ft. 
Spot Spot 


3-Ft. 30-Ft. 
Spot Spot 


9 88 

13 123 
15 150 
20 200 


7 66 

9 87 
10+ 100 
13 125 


Side Arcs 


40 


60 Solid Angle at 15 Feet 


26 23 



The long- wave infrared, band C, evidences itself in exactly the same 
way as heat from a fireplace. It does not penetrate water, but is ab- 
sorbed by it. 

The extreme limits of these three spectral bands are fixed for our 
purposes by the properties of the glass used in the lamp houses and in 
the camera lenses. Glass of the kinds used for such purposes cuts off 
radiations of wavelengths less than 3400 A units in the ultraviolet, and 
of wavelengths greater than 50,000 A in the infrared. The radiant 
energy from the sun when it reaches the earth's surface contains no 
wavelengths shorter than about 2950 A, and none longer than 
50,000 A. 

For the purposes of motion picture photography the only band of 
any value is A, the photographically effective, and the two infrared 



Nov., 1935J 



ENERGY FROM LIGHT-SOURCES 



387 













<N 


TH 




00 




36 


g2 


i> os 

co oo 


OS l> 


CO 


t> TH 


t>. to 


CO l> 

TH O 

tO <M 


fO ^ 

co co 


i 


Hfi 


?is 


1> rH 


CO rH 


CO rH 


rH 6 


<N O 


c<i o 


CO O 





1 lie 

bfl*i* 'O 


OS co 
l> OS 
^^ OS 


o to 
to co 


^8 


oo o 

CO CO 


rH 00 


si 


CO 8 

OS rH 


00 O 
CO CO 
C^l rH 


1 




Th CO 


<N 


<M 


<N 





o o 





rH 





| |,H 


ii 


S2 SS 

^51 CO 


O to 


t3 ^ 


S 


tO CO 
CO O 


^os 


co 
TH 


1 


1 s'iiSJ 


00 I> 


(N 


(N O 


TH 


o 6 


o 6 





i-H 


o 


GO PL( "^ 




















t> of|^ 


co^ 


Sg 


to 


tO^ 


goo 


<N CO 


S?2 


s 


g 


"S j^!3 &o 


tO rH 


t> CD 


T^H CO 


O to 


rf( O 




l> O 


O rH 


CO 


j "IT 8 


OS 00 


<M 


(N 


(M O 





O O 





rH 


o 


5 


















" 


1 111 

:sWij 




Sg 


ss 


28 


"3 


% 


w 


"g 


10 

TH 

o 


j 


















8 


^ 1 

I > 




3 


FS 


8 


g 


8 


to 
to 


TH 

CO 


CO 























i ! 




TH 


TH 





O 


8 


o 


o 


o 























*, 




















1 


1 
















1 


^ 


















3 


! i 


A 

r*\ O 

II 


. I. Carbons 






f 

CO 








1 


hq 




HI 

ll 

Is 






W cT 

Is 








a M 

*flj 

ssl 

a 06 a 



388 F. T. BOWDITCH AND A. C. DOWNES 

bands should therefore be kept as low in intensity as possible. The 
carbon arc sources shown here have amounts of the infrared per unit 
of useful radiation as small as, or smaller than, any other artificial 
sources of illumination. 

Fig. 2 shows that with the exception of the 10-foot spot from the 
sun arcs, the radiant energy received by an actor working under these 
light-sources is only a small fraction of the energy of direct sunlight. 
This is true not only of the total energy but also of any of the three 
bands into which the radiant energy has been divided. Even in the 
case of the most powerful sun arcs, in a 10-foot spot, the total radiation 
is only a third that of sunlight. 

Table I shows the number of units of these various carbon arc 
studio lighting units that would be required to give intensities, both 
total and photographically effective, equal to that of June sunlight. 
For example, the table shows that thirteen 70-ampere rotary spot 
lamps would have to be concentrated upon a 3-foot spot to equal the 
total radiant energy from the sun, and that twenty of them would 
have to be concentrated upon the same spot to equal the photo- 
graphically effective energy from the sun. 

Table II shows the amounts of energy in gram calories per square 
foot per second which these carbon arc sources give under various 
conditions of current and voltage. 

These data show that it would require a very large number of artifi- 
cial light-sources to approximate the heating effects of sunshine in 
Southern California. The various types of arcs expose the actor to a 
total radiation varying in intensity from a maximum of one-third, in 
the case of the sun arc, to less than one-hundredth that of June sun- 
light per lamp. 

REFERENCE 

1 JOY, D. B., AND DOWNES, A. C.: "A New Alternating Current Projection 
Arc," J. Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 116. 



MODERN INSTRUMENTS FOR ACOUSTICAL STUDIES* 
E. C. WENTE** 

Summary. Up to the time of the development of the telephone current amplifier, 
measurements in sound were for the most part made with instruments operating on 
mechanical principles. At the present time almost all such measurements are made 
by means of microphones, amplifiers, and other specially designed electrical instru- 
ments. This paper gives a descriptive survey of various recently developed electrical 
devices used in acoustical studies and a brief discussion of their limitations and their 
field of application. 

Instruments for acoustical research should demand particular con- 
sideration upon the part of the motion picture engineer if for no 
other reason than the fact that it was the development of such devices 
that made possible the change from the silent to the sound pictures. 
The condenser microphone, in the form in which it was almost ex- 
clusively used for the production of the first commercially successful 
sound pictures, was originally developed for the study of sound- 
waves and extensively used for this purpose before it found any 
commercial application. The light-valve and the oscillograph, as 
used for the first variable-width sound pictures, were both originally 
conceived and developed as instruments to be used for scientific 
investigations in acoustics. 

The application of special instruments for the analytical study of 
sound began during the last century. One of the first and outstand- 
ing of these instruments, although never used extensively, was the 
phonautograph of Scott (1857). This instrument consisted essen- 
tially of a diaphragm, to the center of which was attached a stylus 
resting upon a moving strip of smoked paper. When the diaphragm 
was set in motion under the action of sound-waves, a wavy line was 
traced upon the paper, giving a graphical picture of the pressure 
variations of the sound-wave. This device is of special interest as it 
appears to have been the first instrument to comprise a diaphragm of 
the general form found in most present-day acoustical instruments and 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Bell Telephone Laboratories, Inc., New York, N. Y. 



390 E. C. WENTE [ j. s. M. p. E. 

in two of the outstanding inventions of the last century, the telephone 
and the phonograph. The phonautograph, when compared with 
Edison's first phonograph, shows how a purely scientific instrument 
can be the precursor of a great invention. 

During the last century one of the chief problems in acoustical 
measurements was to devise instruments capable of a measurable 
response to the exceedingly small pressures that obtain in ordinary 
sound-waves. Laboratory and commercial instruments generally 
had to depend upon resonance or sound concentrators for their sensi- 
tivity, both of which, according to present standards, introduce large 
amounts of distortion. About twenty years ago, however, came the 
development of the telephone amplifier through the invention of the 
audion by deForest. In conjunction with a microphone, such an 
amplifier could translate the small power of a sound-wave into corre- 
sponding electrical power great enough to operate electrical meters 
or recorders of various kinds. It was no longer necessary to resort to 
devices that introduced distortion in order to attain the necessary 
sensitivity. Soon after these amplifiers became available, micro- 
phones were therefore devised for the study of sound-waves of 
ordinary intensities which introduced distortion of a much lower 
order of magnitude than the acoustic instruments previously avail- 
able. Instruments in which a microphone is an essential element 
have so come to dominate the field of acoustic measurements that we 
may disregard all others hi our present discussion. 

In one sense the microphone is really the only acoustic instrument 
used in most sound measurements, for after the sound-wave has been 
translated into a corresponding electrical current we need only apply 
the tools of the electrical engineer for measurement or analysis of this 
current to determine desired characteristics of the sound-wave. We 
shall, therefore, first give some consideration to the microphone, and 
follow this with a discussion of some recently developed electrical 
instruments that are particularly useful in the study of acoustical 
problems. 

Microphones. A sound-wave is characterized by periodic changes 
of pressure, temperature, density, and particle velocity. A device 
that translates any one of these characteristics into a corresponding 
electrical current is a microphone. Microphones of each of these 
types have been proposed by various investigators. Pressure, tem- 
perature, and density are scalar quantities proportional to each other 
at any point in a sound field, so that microphones responding to any 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 



391 



one of these characteristics may be conveniently grouped under the 
general heading of pressure microphones. However, velocity is a 
vector quantity, the magnitude of which is not always proportional 
to the pressure in the sound-wave. So we shall expect to find charac- 
teristic differences in the response of pressure and velocity micro- 
phones, depending upon where and how they are placed in a sound 
field. Which of these two classes or what combination of the two is 
to be preferred depends upon the nature of the problem under in- 
vestigation. 

If the microphones are assumed to be small compared with the 
wavelength, so that they will not distort the sound field by diffraction, 
then the response of a pressure microphone will be independent of the 
direction of propagation of the sound-wave; whereas, that of the 




100 500 1000 

FREQUENCY IN CYCLES PER SECOND 

FIG. 1. Response vs. frequency characteristic of a velocity 
microphone, for various distances from a source of constant 
strength. 

velocity microphone will be proportional to the cosine of the angle 
between this direction and a line drawn normal to the face of the 
microphone. Hence, if measurements are to be made of sound com- 
ing from a known direction, the velocity microphone can often be so 
oriented as to discriminate against interfering sound-waves. The 
application of this feature in various kinds of acoustic measurements 
has been discussed in a recent paper by Wolff and Massa. 1 On the 
other hand, results of measurements in complex sound fields can often 
be more easily interpreted when obtained with a pressure micro- 
phone, because of its non-directional characteristics. 

The reactions, in respect to the sound field, of pressure and of 
velocity microphones differ in another important respect because of 
the fact that in the sound field of a simple source the pressure varies 
inversely with the distance from the source at all frequencies ; but the 
velocity increases more rapidly at low than at high frequencies as the 
distance from the source is decreased. If, then, the source emits 



392 E. C. WENTE ( j. S. M. P. E. 

sound of complex wave-form, the pressure microphone will deliver a 
voltage of identical form, regardless of its position in the sound field; 
but the velocity microphone will generate a voltage with a greater 
proportion of low-frequency components at near than at far points. 
The magnitude of this change is indicated by the response vs. fre- 
quency characteristics, shown in Fig. 1, for various distances from a 
source of constant strength, a flat characteristic being assumed for 
the instrument when placed at a great distance from the source. 
This variation, while of no great practical consequence when the 
instrument is used at moderate distances, must be taken into account 
in measurements made near the source, and exacts a careful control 
of placement. 

It has been pointed out that with the advent of amplifiers, sensi- 
tivity in a microphone became of secondary importance. It is not, 
however, a matter to be disregarded entirely, because there is a prac- 
tical lower limit to the voltage that can be amplified, and conse- 
quently for a microphone of a given sensitivity there is a lower 
limit to the sound intensity at which the microphone can be usefully 
employed. This ultimate limit to amplification is set by the noise 
voltage generated by the thermal agitation of the electrons in the 
input resistance of the amplifier. 2 The most sensitive high-quality 
microphone so far described is of the moving-coil type, similar to the 
Western Electric Company's 618 microphone. It is, therefore, of 
interest to see over what range of intensity sound can be successfully 
translated by this instrument. A potential of about 10 ~ 4 volt is 
developed per bar of sound pressure, and its electrical resistance is 
approximately 20 ohms. The voltage due to thermal agitation of the 
20-ohm resistance over a 15,000-cycle band is equal to 7 X 10~ 8 . 
Assuming that the microphone can be usefully employed down to 
potentials equal to the thermal potential, we find that the lowest 
measurable sound pressure is 7 X 10 ~ 4 bar. Lower values of pres- 
sure can be measured if the transmitted frequency band is decreased, 
because the noise power is directly proportional to the width of this 
band. The two lower curves of Fig. 2 show graphically the pressures 
in a plane wave that can be measured at various frequencies for a 
15,000- and a 200-cycle band, account being taken of the variation in 
sensitivity of the instrument with frequency. The shaded area 
gives the range of hearing. It is seen that even with this relatively 
sensitive microphone in the region where the ear is most sensitive, 
the sound pressures must be some 15 decibels above the threshold of 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 



393 



audibility if the instrument is operated over the wider band. This 
microphone, when placed in a plane sound field, will translate approxi- 
mately 1 per cent of the sound falling upon the diaphragm into elec- 
trical power. It is interesting to note that if this efficiency were 
raised to 100 per cent, the lowest pressures that could be measured in 
this region would be just about equal to what the ear can hear. As 
the sound pressure is raised to higher levels, the microphone will 
generate harmonic voltages of an increasing amount. The upper 
curve of the diagram shows the pressures at which the harmonic 
components are 30 decibels below the fundamental. Above this 
level the harmonic components begin to increase rapidly. Over 




-20 



0.00002 



20 50 100 500 1000 5000 10,000 20,000 

FREQUENCY IN CYCLES PER SECOND 

FIG. 2. Showing the pressures in a plane wave that can be measured at 
various frequencies, for 15,000- and 200-cycle transmitted bands. 

most of the frequency range the operating pressures are considerably 
above the feeling level for the ear, which is represented by the upper 
boundary of the shaded area. However, many sounds encountered 
in civilized life that we may wish to study lie above the feeling level. 
Rapid Record Oscillograph. If we wish to make a detailed study 
of the properties of the sound-wave not having a steady character, 
theoretically the most perfect method available for the purpose is to 
use an oscillograph, which translates the time pattern of the micro- 
phone current into a corresponding space pattern. If the oscillo- 
graph is to give a true graphical representation of the instantaneous 
pressures in the sound-wave, i. e., an accurate picture of the wave- 
form, it is necessary not only that its sensitivity be the same at all 
frequencies, but that it preserve the phase relationships as well, a 



394 



E. C. WENTE 



[ J. S. M. P. E. 



condition that need not be satisfied in a record to be used solely for the 
reproduction of sound. In the past, conclusions have frequently been 
erroneously drawn regarding the characteristics of certain types of 
sound-waves upon the basis of records obtained with oscillographs not 
free from phase distortion. The appearance of the record can be 
completely altered by a shift in the relative phases of the components. 
In acoustical problems, when we wish to study the effect of various 



PAPER 
ROLL 



CLUTCH ROLL- 
CYLINDRICAL LENS- 
VIEWING SCREED 




STRING 
CONDENSING 

OCCULTING^ EN v 
TIMING 
WHEEL 



LAMP 



STORAGE TANK 



FIG. 3. Diagram of the Curtis string oscillograph. 



conditions upon the sound-wave, it is a great convenience, and some- 
times essential, that the record be available for inspection as soon as 
possible after the sound has been recorded. An instrument satisfying 
these conditions is the Curtis string oscillograph, 3 the operation of 
which is shown diagrammatically in Fig. 3. This instrument is free 
from both frequency and phase distortion up to about 8000 cps. A 
strip of photographically sensitive paper is passed by the point of ex- 
posure into a reservoir from which it is fed through processing baths. 
The record is available for inspection about one minute after exposure. 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 



395 



This instrument has been found to be extremely useful in many kinds 
of acoustic investigations. 

When the sound to be studied is steady, the wave-form may be 
observed with a cathode ray oscillograph, which is available in many 
forms. By the use of a sweep circuit of the type described by Bedell 
and Reich, 4 the repeated cycle may be automatically stabilized into a 
stationary pattern. 

Harmonic Analyzers for Steady Currents. In many acoustic 
measurements it is not so important to know the exact form of a 
sound-wave as its harmonic components. It is, of course, possible to 
determine these components from an oscillogram by well-known 
mathematical or instrumental methods. These methods, except 
where only a few components are desired, are slow and laborious. 
Components that are small compared with the largest components 




2 3 4 56789 10 II 

FREQUENCY IN KILOCYCLES PER SECOND 

FIG. 4. Analysis of a steady complex sound. 

can usually not be determined with great accuracy from oscillograms. 
When the current to be analyzed has a steady character, one of a 
number of so-called current analyzers may be used. These have been 
developed in several different forms, but have the common feature of 
comprising a narrow band-pass circuit, either mechanical or electri- 
cal, the mid-frequency of which is variable continuously or in small 
steps over the frequency range. The analysis is made by varying the 
frequency of this transmitting circuit and reading the corresponding 
transmitted currents upon a meter, or registering them with an 
appropriate recorder. The narrower the frequency band of the trans- 
mitting circuit the greater the degree to which closely spaced com- 
ponents may be distinguished. The wider band is, however, to be 
preferred where detailed resolution is not required, as the analysis 
may be made at a proportionately greater speed. An analyzer of this 
type, in which the width of the transmitted frequency band may be 
set to either 20 or 200 cycles, was described by Wolf and Sette. 5 The 
filters in this analyzer have sharp discriminations at the edges of the 



396 



E. C. WENTE 



[J. S. M. P. E. 



transmitting band, so that even closely spaced components may be 
measured with a high order of precision. 

Fig. 4 shows the type and form of the results that may be obtained 
with this analyzer when used in conjunction with a level recorder. 
The particular record shows the harmonic components in the current 
of an audiometer of the buzzer type. It gives the relative intensities 
of about 70 components, the amplitudes of which cover a range of 
more than 50 decibels. This type of instrument is also useful in the 
study of the statistical distribution with respect to frequency of the 
components in a sound that is not steady in character. Fig. 5, for 
example, shows the result of an analysis of the sound emitted by a 
small electric drill. 

High-Speed Analyzer. Analyzers of the type just described are 



jJ lOOf 

I 60i 



Z 60 



20-CYCLE BAND 







34 567 

FREQUENCY IN KILOCYCLES PER SECOND 



FIG. 5. Analysis of the sound emitted by a small electric drill. 

particularly suitable for the analysis of steady-state currents. Many 
sounds, however, assume a steady-state value for only a short period. 
Unless a statistical value is desired, the analysis must then be made 
by sweeping through the frequency range during this period. Other 
kinds of sounds, such as speech or music, vary in their harmonic 
composition almost continuously. At first thought we should con- 
clude that such changes in composition might be followed by sweeping 
the transmitting frequency band back and forth rapidly, and record- 
ing the transmission throughout each sweep cycle with a high-speed 
recorder. It is possible to proceed a certain distance in this direction ; 
but no matter what type of transmitting circuit is used, it will have 
a finite time-constant; i. e., a finite time is required for the current to 
reach a certain fraction of its steady-state value when the circuit is 
set to a given frequency. Similarly, a finite time interval is required 
for the current to die down to a certain fraction of the steady-state 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 



397 



value when the frequency is changed. For this reason, as the rate of 
shift of the frequency band is increased, the precision with which the 
harmonic components are determined is decreased. Only a very 



CYCLES PER SECOND 
20O 400 800 



I | r i-..|.-|"H 

PIANO TONE F| 

|*.~". I I'"" 

PIANO TONE F 

| I I |'.||...| |..|..|,i.,,.|.i.| 

OPEN a SUNG AT PITCH OF Gg 

|' I 'I | ll-| "'"I I 

NASAL S SUNG AT PITCH OF Gg 

I I |l ill - II II 

OPEN a SUNG AT PITCH OF C 

" ' | |l l-i-l.-l | 

OPEN SUNG AT PITCH OF G 

'I I' 



NASAL 5 SUNG AT PITCH OF C, 2 



FIG. 6. Records of typical sound obtained with the Hickman analyzer. 

rough analysis of the waves of ordinary speech or music will be 
possible by this method. The speed of the analyzer can be greatly 
increased without sacrificing resolution by employing, instead of a 
single transmitting element of variable transmission frequency, a 



398 E. C. WENTE [j. S. M. P. E. 

large number of elements closely spaced in frequency and all opera- 
tive simultaneously, for in this case the current needs to be main- 
tained only sufficiently long for the slowest element to approach its 
steady-state value. 

An analyzer built up from electrical filters of the required number 
would be both expensive and bulky. A simple instrument of this 
type, having mechanical resonance elements simultaneously opera- 
tive, has been developed by Hickman. 6 A series of tuned reeds 
with their resonance frequencies spaced at equal pitch intervals are 
simultaneously electromagnetically driven by the current to be 
analyzed. These reeds are especially designed so that each one will 
be set in resonant vibration at only one particular frequency. A 
mirror is attached to each reed from which light is reflected and 
brought to focus upon a screen in the form of a short fine line. The 




' FIG. 7. Curve obtained with the sound decay 
high-speed level recorder. 

broadening of this line as the mirror vibrates is proportional to the 
amplitude of motion of the reed, which, in turn, is proportional to the 
strength of the component in the driving current having a frequency 
approximately equal to the resonance frequency of the reed. The 
amplitude of the various components may be thus observed simul- 
taneously upon the screen or the changes in the composition with time 
may be recorded by a motion picture camera. Fig. 6 shows the form 
in which records of some typical sounds are obtained with this in- 
strument. 

Another novel form of high-speed analyzer has recently been 
described by Meyer. 7 By a method well known in communication 
engineering, the frequencies of the various harmonic components of 
the current are increased by equal amounts. After this the current 
is translated into sound by a special high-frequency loud speaker. 
The components of the radiated sound are then all of short wave- 
length, which are reflected from a grating consisting of a row of rods 
equally spaced along a concave surface. Analogously to light re- 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 



399 



fleeted from a concave ruled grating, the components of the sound are 
brought to a focus at different points along a focal surface. Along 
this surface a small high-frequency microphone is moved back and 
forth rapidly. The current generated by the microphone along its 
path is then recorded. From the record so obtained the relative 
strength of the various components is determined. 

High-Speed Level Recorder. In some important types of sound 
measurements we are not interested in a detailed analysis of the 
sound-wave but merely in the variation with time of the average level 




FIG. 8. Variations of level of sound from a piano, obtained with the 
high-speed level recorder. 

of the sound, as in the measurement of the rate of decay in a room or 
the flow of energy in speech, music, or noise. In some cases this 
average is preferably taken over long and, in others over short, time 
intervals. For long-time averages, a thermocouple or rectifier and an 
ammeter may be used, but for short-time averages an instrument is 
required that can follow changes of intensity at a higher rate. Fre- 
quently, also, the range of intensity over which we desire to make 
measurements of this character is very wide. Reverberation mea- 
surements are preferably made over a range of at least 60 decibels, 
and the level range of orchestral music covers about 75 decibels. 
Several instruments designed for such purposes have been described 
recently. 8 In the instrument described by Wente, Bedell, and 



400 



E. C. WENTE 



[ J. S. M. P. E. 



Swartzel the level is recorded by a stylus upon waxed paper. The 
recorder can be adjusted to give either a short or a long time average. 
At the highest operating speed it is capable of following changes of 
intensity at the rate of 840 decibels per second, and fluctuations in 
intensity of about 100 per second. The instrument may be adjusted 
so that the full scale covers a range of 30, 60, or 90 decibels. 

Fig. 7 shows the type of curve obtained with this recorder in the 
measurement of the rate of decay of sound in rooms. Fig. 8 illus- 




100 500 1000 

FREQUENCY IN CYCLES PER SECOND 



FIG. 9. Relation between aural stimulus and sensation, the various 
curves giving the intensity levels of pure tones of equal loudness. 

trates its use as a recording volume indicator. It shows the varia- 
tions in level during the playing of a phrase upon the piano. The 
curves from top to bottom were obtained with the recorder set to 
operate at increasingly higher speeds. From records of this type the 
dynamics of the performance of a musical selection and the range of 
level can be accurately checked and studied. 

Measurement of Pitch. For acoustical studies, where it is of no 
particular importance to know the wave-form, but interest lies in the 
variation of pitch with time, as in the study of the vibrato in musical 
tones, or in the inflections of the speaking voice, several types of 
instruments have been devised. Perhaps of these the most widely 



Nov., 1935] INSTRUMENTS FOR ACOUSTICAL STUDIES 401 

known is the tonoscope, developed by Seashore and his associates, 
which operates on the stroboscopic principle. This instrument has 
rows of uniformly spaced dots upon a rotating cylinder, the number of 
dots increasing in successive rows. A neon light is made to flicker in 
synchronism with the fundamental of the tone under investigation. 
The particular row which under the light appears stationary gives 
the pitch of the tone at any instant. By the aid of a suitable camera, 
the time variations of pitch may be recorded photographically, giving 
a so-called strobophotograph. 

A frequency recorder operating on a different principle has been 
described by Hunt. By a special circuit arrangement, employing 
gas-filled discharge tubes in combination with a spark recorder, the 
pitch of a tone can be recorded upon paper. The scale is linear up to 
8000 cycles. This instrument is capable of following changes in 
pitch at a high rate. 

Loudness and Its Measurement. The preceding discussion was 
restricted to the purely objective or the physical aspects of sound. 
Often we are finally interested in not the physical nature of sound but 
its subjective characteristics as perceived by our ears. The relation- 
ship between stimulus and sensation is very complex, as may be 
observed from the curves shown in Fig. 9. The various curves give 
the intensity level of pure tones of equal loudness. The threshold 
of audibility varies widely with frequency, and the relationship 
between sensation level and intensity level is not the same at the 
various frequencies and levels ; for instance, at a loudness level of 40 
decibels above threshold, a change of 5 decibels in the stimulus at 
100 cycles produces the same change in sensation as a change of 10 
decibels at 1000 cycles. For complex tones the relationships are 
often more complex, although Fletcher and Munson have developed 
formulas whereby the loudness level of a steady sound can in most 
cases be computed from the intensity level of its components. In 
spite of the above apparently complex relationships, so-called noise 
or sound meters have been developed that indicate upon a scale the 
subjective intensity. These meters have proved themselves particu- 
larly useful in the measurement and study of noise. 



REFERENCES 

1 WOLFF, I., AND MASSA, F. : "Use of Pressure Gradient Microphone for Acous- 
tical Measurements," J. Acoust. Soc. Amer. t 4 (Jan., 1933), No. 3, p. 217. 



402 E. C. WENTE 

'JOHNSON, J. B.: "Thermal Agitation of Electricity in Conductors," Phys. 
Rev., 32 (July, 1928), No. 1, p. 97. 

'CURTIS, A. M.: "An Oscillograph for 10,000 Cycles," Bell Syst. Tech. J., 
XH (Jan., 1933), No. 1, p. 76. 

4 BEDELL, F., AND REICH, H. J. : "Cathode Ray Oscillograph for Several Simul- 
taneous Waves to Stabilize Linear Time Axis," Science, 63 (June, 1926), No. 1642, 
p. 619. 

6 WOLF, S. K., AND SETTE, W. J.: "Some Applications of Modern Acoustic 
Apparatus," /. Acoust. Soc. Amer., 6 (Jan., 1935), No. 3, p. 160. 

HICKMAN, C. N.: "An Acoustic Spectrometer," J. Acoust. Soc. Amer., 6 
(Oct., 1934), No. 2, p. 108. 

7 MEYER, E., AND THIENHAUS, E.: "Schallspektroskopie, ein neues Verfahren 
der Klanganalyse," Zeitschr.fur Tech. Phys., 15 (1934), No. 12, p. 630. 

8 HUNT, F. V.: "Recording Instruments for Frequency and Intensity," 
/. Acoust. Soc. Amer., 6 (July, 1934), No. 1, p. 54. 

WENTE, E. C., BEDELL, E. H., AND SWARTZEL, K.: "A High Speed Level- 
Recorder for Acoustic Measurements," /. Acoust. Soc. Amer., 6 (Jan., 1935), 
No. 3, p. 121. 



FLUTTER IN SOUND RECORDS* 

T. E. SHEA, W. A. MACNAIR, AND V. SUBRIZI** 

Summary. Frequency modulation of a sound signal is caused by non-uniformity 
in the record speed during the recording or reproducing process. This source of 
flutter is discussed and was demonstrated at the May, 1935, Convention. 

The paper includes a discussion of the physical nature of frequency modulation, the 
physiological effects of frequency modulation, the methods of producing known 
amounts of artificial flutter , and the methods of measuring flutter . 

There are various forms of distortion which may degrade the 
quality of sound signals. Many of these types of distortion are 
well known, and their effects readily identified by ear by the ex- 
perienced engineer. For example, the effect of a limited volume range 
is characterized by a high noise level. Also the effect of a limited 
frequency range has been demonstrated many times, * and engineers 
quickly detect the lack of crispness and naturalness of certain types 
of sounds when the higher frequencies are suppressed. Furthermore, 
the effect of non-linear response of a portion of the electrical circuit, 
or "overload," is well known. 

Some other forms of distortion are more obscure, and perhaps are 
not readily detected by as many persons. Frequency modulation^ 
of a sound signal produces some very interesting and sometimes not 
readily identifiable auditory sensations. Distortion of this type is 
caused in sound records by the non-uniform speed of the record 
during the recording or reproducing processes. An analysis of the 
distortion leads to the location of the mechanical cause of the non- 
uniform speed, and consequently leads to improved record-driving 
mechanisms. 

The methods of quantitative analysis of these speed variations 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Bell Telephone Laboratories, New York, N. Y. 

t E. g., at the meeting of the Atlantic Coast Section, Jan. 30, 1934, at New 
York, by H. Fletcher. 

tt Where the word "flutter" is used, its meaning will be restricted to the dis- 
tortion due to frequency modulation. 

403 



404 SHEA, MACNAIR, AND SUBRIZI [J. S. M. P. E. 

will be described, and the physical nature and some of the physio- 
logical effects of the distortion caused by them will be discussed 
and demonstrated. 

PHYSICAL NATURE OF FREQUENCY MODULATION 

Frequency modulation 1 consists of a cyclic change in the frequency 
of a recurrent wave, the cyclic amplitudes of the wave remaining 
unchanged. It is equivalent to a distortion of the time-axis scale 
of a recurrent wave or a periodic variation of the time-delay of the 




FIG. 1. Frequency-modulated and amplitude-modulated oscillator tones. 

(A) Frequency-modulated oscillator tone. 

(B) Constant-frequency oscillator tone. 

(C) Amplitude-modulated oscillator tone. 

(D) Constant-frequency oscillator tone. 

signal. If one puts a constant-frequency record into a reproducing 
machine that has a periodic speed variation, the output signal has 
the form shown in an exaggerated way in Fig. 1. In the oscillogram 
marked A there are periodic changes in frequency. At the time 
when the wave cycles were near together, the reproducing machine 
was running above average speed; and at the time when the waves 
were spread out, the reproducing machine was running at lower 
than average speed. This periodic speed change of the reproducing 
machine, which gives a signal of the type illustrated here, is "nutter." 
The nutter in the signal is, technically, "frequency modulation/' 
because the frequency of the signal periodically swings above and 



Nov., 1935] FLUTTER IN SOUND RECORDS 405 

below normal. For comparison, oscillogram B, of a constant 
frequency tone, is included. In contrast, an amplitude-modulated 
wave is shown in oscillogram C, which is the familiar type of modu- 
lation used on the radio broadcasting channels. 

Frequency modulation may be very complex in character. Where, 
however, a single-frequency tone is considered and the cyclic change 
in frequency is itself sinusoidal, the amount and character of the 
distortion are described by three factors : 

(1) The frequency of the normal tone (/o). 

(2) The extent of the cyclic change in this frequency ( A/ ). 

(3) The frequency of the cyclic change (f m ). 

The extent of the cyclic change is usually expressed in percentage 
form (A/O//O X 100); this may be called the amplitude of the dis- 
tortion. The frequency f m may be called the rate of the distortion. 
We shall use the terms "amplitude" and "rate" to express the char- 
acter of various frequency modulations. 

Frequency modulation of the sinusoidal type may be shown to be 
physically equivalent to (1) the production of side-frequencies, 
spaced at uniform discrete intervals; and (2) a reduction in the 
amplitude of the normal or "carrier" tone. The frequency interval 
between the sideband components is equal to the rate of the modula- 
tion, and the magnitudes of the various components depend upon the 
value of the ratio (A/ // w ). Various analyses of frequency modu- 
lation have been published. 2 

The instantaneous amplitude of any undistorted sinusoidal wave 
of frequency f may be written : 

y = A sin w t (1) 

where the condition co = 2irf is dependent upon the maintenance 
of uniform frequency. If the frequency varies, co must be regarded 
as a variable; i. e., 

y = A sin I wdt (2) 

If the variation in the frequency is sinusoidal then : 

co = co + Ao sin u m t (3) 

where <o m = 2wf m . 



406 SHEA, MACNAIR, AND SUBRIZI [J. S M. p. E. 

Substituting (3) in (2), 

y = A sin I (co + Aco sin co t)dt 
J o 

r A< * /v 

= A sin I co f cos w m t I 

L co Jo 

, . Acoo Acoo 
= A sin o) t -\ cos CO OT t (4) 

L C0 m CO OT J 

The second term within the brackets of (4), namely, AOO O /CO OT , is a 
constant phase-angle, and for simplicity may be dropped out; whence 

y = A sin (<a t a cos CO TO /) (5) 

and 

y = A [sin co J cos (a cos u m t) cos co / sin (a cos co m O ] (6) 

The well-known Fourier developments for cos (x cos 3;) and sin 
(x cos y) are : 3 

cos (x cos y) = J (x) 2Jz(x) cos 2y + 2/ 4 (#) cos 4y . . . (7) 

sin (x cos y) = 2Ji(x) cos y 2/ 3 (*0 cos 3 y + ... (8) 

in which the J's are the well-known Bessel's coefficients. Expanding 
(6) in accordance with (7) and (8). 



y = A [sin co ^{/ (a:) 2/ 2 (a) cos 2co m t + 2/ 4 (a) cos 

cos aW{2/i(a) cos CO TO / 2/ 3 (a) cos 3co w / + .}] (9) 

which may be written 

y = AJ (a) sin w t carrier 

A Ji (a) cos(co + co m )J] 1st upper side-frequency 

A J\ (a) cos(co<> co TO )J] 1st lower side-frequency 



+ 
+ 



sin( 



co 



sm(ou<, 
cos(co -f 

COS(C0 



2nd upper side-frequency (10) 
2nd lower side-frequency 
3rd upper side-frequency 
3rd lower side-frequency 



The magnitudes of the carrier and the various side-frequencies 
relative to that of the original undistorted wave are given by the 
coefficients /<,(<*)> & The locations of the side-frequencies are 
given by the values of co co m , etc. 

From tables of Bessel's functions, we may plot the values of the 
/'s as a function of a as in Fig. 2. Some typical examples illustrate 
the use of Fig. 2. Thus, if a 2000-cycle tone is modulated 5 per cent 
(from 2100 cycles to 1900 cycles) 20 times per second, a - A/ // w 



FLUTTER IN SOUND RECORDS 



407 



Nov., 1935] 

= 100/20 = 5.0, the relative magnitudes of the carrier and the suc- 
cessive orders of side-frequencies are (Fig. 3, lower half) : 

0.17, 0.32, 0.04, 0.36, 0.39, etc. 

The components are spaced 20 cycles apart. The carrier is small 



i.o 



0.9 



UJ 
= 0.7 

UJ 

a 
u.0.6 



\ 



CARRIER 



\ 



NUMBERS REFER TO SUCCESSIVE 
SIDE FREQUENCIES 



0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 




0.5 



FIG. 2. Relation between carrier and side-frequencies in frequency- 
modulated tones. 

in this case, but a number of the side-frequencies are substantial in 
magnitude. Obviously, there are values of a for which either the 
carrier or any specific order of side-frequencies may be wiped out. 
On the other hand, where a is small (due either to a small departure 
from average frequency or rapidity in this fluctuation), the side-band 
components are small and relatively unimportant, while the carrier 



408 



SHEA, MACNAIR, AND SUBRIZI 



[J. S. M. p. E. 



has nearly its original magnitude. If a = 0.25 (upper half of Fig. 3), 
the carrier has a value of 0.98 and only the first-order components, 
with a value of 0.13, are of appreciable magnitude. 

Physically, then, distortion due to frequency modulation is re- 
markably like non-linear distortion in that extraneous frequencies 
are produced having a systematic relation to the fundamental or 



0.6 



0.4 



gO.2 



CARRIER FREQUENCY 



FREQUENCY MODULATED 
0.25 % AT 20~ 



tt = 0.25 



0.6 



0.4 



0.2 



CARRIER FREQUENCY 

= 2000^ 

FREQUENCY MODULATED 
5% AT 20^ 



a =5.0 



1800 1840 1880 



1920 I960 20OO 2040 2080 
FREQUENCY IN CYCLES PER StOOblD 



2120 2160 2200 



FIG. 3. Acoustic spectra of certain flutter tones. 

normal tone. The physiological effects of this distortion will be 
discussed below. 



PHYSIOLOGICAL EFFECTS OF FREQUENCY MODULATION 

It is a part of this paper to demonstrate the aural effects of fre- 
quency modulation of sound signals. The effects produced, of 
course, will vary with amplitude and rate of the modulation, but will 
also vary greatly with the acoustical listening conditions. 

Hanson 4 has demonstrated that a pure tone generated in a com- 



Nov., 1935] FLUTTER IN SOUND RECORDS 409 

paratively "dead" room sounds the same as the same tone generated 
in a "live" room. If, however, the signal is frequency-modulated, 
then the character of the sounds picked up in the two rooms is quite 
different. Oscillographic analysis of the sounds as picked up by a 
microphone shows that the sound from the "dead" room is nearly 
the same as the signal supplied to the loud speaker; that is, of uni- 
form amplitude, but modulated frequency. The signal picked up 
from the "live" room, however, exhibits very marked amplitude 
fluctuations of complicated character, indicating that the "live" 
room has operated upon the frequency-modulated signal in such a way 
as definitely to change its character. This is illustrated in Fig. 4, 
in which the signal from the "dead" room is of nearly uniform ampli- 
tude, but the signal from the "live" room shows large fluctuations 



second- 



FIG. 4. Frequency-modulated tone in "live" and "dead" rooms. 
(A) Frequency-modulated tone in "live" room. 
(J5) Frequency-modulated tone in "dead" room. 

in the amplitude of the signal. It is to be expected, therefore, that 
the frequency-modulated signal will sound quite differently when 
heard in a "live" room than in a "dead" one. 

Under any given listening condition there is a marked difference 
in the aural sensation produced by a 1000-cycle tone modulated at 
rates of about 40 cycles and above, and by the same tone modulated 
at about 20 cycles and below. At the latter rates of modulation one 
hears a tone that sweeps up and down in pitch across the frequency 
band. That is, one hears the modulated signal in the form that 
would be the natural interpretation of equation (5) and the dis- 
cussion leading up to it. For rates of modulation of about 40 cycles 
and above, however, one hears a group of frequencies that are not 
harmonically related to one another and which impart a harsh quality 
to the tone. In this case, one hears the modulated signal in the 
form that would be the natural interpretation of equation (10). 
Now, it has been shown that, analytically, equations (5) and (10) 



410 



SHEA, MACNAIR, AND SUBRIZI 



[ J. S. M. P. E. 



are merely two ways of writing the same thing. It is interesting that 
(5) appears to be a more natural interpretation for low rates and (10) 
for higher rates of modulation. 

In producing musical tones musicians often use artistic frequency 
modulation to improve the musical quality of their performance. 
An important component part of the vibrato is a frequency modula- 
tion at a rate of about 6 cps., with an amplitude of plus and minus 
several per cent. 

That the auditory effects of a frequency-modulated signal and an 
amplitude-modulated signal may be very similar under certain con- 
ditions is shown in Fig. 5. Oscillogram A shows a frequency- 
modulated signal, the modulation being so small that it can not be 




FIG. 5. Frequency modulation and amplitude modulation in a live room. 

(4) Frequency-modulated input to loud speaker. 

(5) Microphone output. 

( C) Amplitude-modulated input to loud speaker. 

(D) Microphone output. 

seen as such; and oscillogram C shows a signal that is amplitude- 
modulated at the same rate. Each of these signals was supplied to 
a loud speaker in a live room, and oscillograms B and D show the 
signals picked up by a microphone in the room. In this case a live 
room has modified both types of signals so that they are practically 
identical. 



METHODS OF PRODUCING KNOWN AMOUNTS OP ARTIFICIAL FLUTTER 

There are a number of methods for producing known amounts of 
artificial flutter. In the case of constant-frequency signals, a simple 
way is to vary the tuned circuit capacity in one of the high-frequency 



Nov., 1935] 



FLUTTER IN SOUND RECORDS 



411 



circuits of a heterodyne oscillator by means of a rotating variable air 
condenser with plates so shaped and driven at such rates as to give 
the desired modulation. 

In the cases of speech and music, the problem is more difficult. 
Here, one way is to produce the undistorted signal in the form of a 
plane acoustic wave, and swing a pick-up microphone to and fro 
along the direction of travel of the sound. Where the signal is 
generated by a loud speaker, the latter may be vibrated instead of 
the microphone, if more convenient. The resulting signal will then 
be frequency-modulated at the rate of the vibrating microphone or 
speaker, and at an amplitude that may be calculated. Obviously, 
room reflections may provide serious complications to this method, 
especially if used for modulating an original sound signal that may 
be coming from several different directions at once, and therefore 




OSCILLOGRAPH 



SIGNAL 
2000 OR- 
3000~ 





MODULATOR 




BAND-PASS 
FILTER 




AMPLIFIER 


STRINGS 
N0.1 








NO. 2 









FIG. 6. Schematic arrangement of a flutter-measuring circuit. 

be subjected to varying amounts of modulation depending upon its 
angle of incidence with the vibrating microphone. The method is 
better suited to distorting the signal from a loud speaker because the 
speaker may be mounted in a sufficiently dead room in such a manner 
that the effect of reflections may be made negligible. 

Generally speaking, it is easier to modulate a signal that is already 
stored in the form of a record. In this case it is necessary only to 
introduce, mechanically, the desired speed variations in the record. 
An alternative is to drive the record at a uniform speed and to vibrate 
the pick-up device so that the scanning point moves along the record 
at a non-uniform speed. For demonstration purposes the desired 
effect has been attained by operating upon a film record played back 
in a re-recording machine. The film speed relative to the scanning 
light was varied by means of a motor-driven cam, the speed and 
eccentricity of which determined the rate and amplitude, respec- 
tively, of the resulting frequency modulation. 



412 



SHEA, MACNAIR, AND SUBRIZI 

METHODS OF MEASURING FLUTTER 



[J. S. M. P. E. 



In the practical analysis of the flutter caused by speed variations in 
a reproducing machine, the signal from a constant-frequency record 
run in the machine is analyzed for frequency modulation. This is 
done by changing the frequency modulation into amplitude modula- 
tion, and demodulating this modified signal in some common form 
of detecting circuit. 

A very useful form of analyzing circuit and one that is at present 



30 

2 



-300 -250 -200 -150 -100 
BELOW 



-50 50 

MID-BAND 

FREQUENCY 

FREQUENCY IN CYCLES PER SECOND 



100 150 200 250 300 
ABOVE 



FIG. 7. Attenuation-frequency characteristics of the band-pass filter. 

in use at Bell Telephone Laboratories is shown in schematic form in 
Fig. 6. A 2000- or 3000-cycle signal (with side-bands) from the 
machine under test is modulated by an oscillator tone. The differ- 
ence-frequency passes through a band-pass filter at the point marked 1 
upon the frequency characteristic shown in Fig. 7. Here the lower 
side-bands are attenuated and the upper side-bands amplified rela- 
tively to the carrier frequency, resulting in an amplitude-modulated 
wave that is amplified and recorded by one string of a rapid-record 
oscillograph. 5 To insure that any amplitude change that might 
occur in the original signal will be observed, another string of 
the oscillograph is used simultaneously to record this signal directly. 



Nov., 1935] FLUTTER IN SOUND RECORDS 413 

Because of the complicated character of a frequency-modulated 
signal it is laborious to calculate the response of such a circuit to 
frequency modulation of various amplitudes and rates. This is 
especially true since in many cases more than the first-order terms 
must be taken into consideration. It is a more practicable expedient 
to calibrate the equipment, using known amplitudes and rates of 
artificial flutter. 

DEMONSTRATIONS ACCOMPANYING THE PRESENTATION OF THIS PAPER 

An extensive demonstration accompanied the presentation of this paper 
at the Spring, 1935, Meeting at Hollywood, Calif., May 20-25, 1935. The 
mathematical portion of the paper was not presented, but, instead, the side-band 
nature of frequency modulation was demonstrated. 

The demonstrations were designed to illustrate the many interesting and 
varied acoustical phenomena that occur in connection with sound flutter, and 
were built up from simple physical phenomena with which most motion picture 
engineers are acquainted. It was not intended to include material relative to 
commercial tolerances in flutter. In order that the phenomena involved could 
be clearly perceived by the audience, the amounts of flutter demonstrated gener- 
ally exceeded greatly what is permissible in commercial practice. 

(1) Visual vs. Auditory Perception of Flutter. Oscillograms and fluttered 
tones were used to show that flutter could be perceived visually in oscillograms 
only when extreme in amount. 

(2) Side-Band Nature of Flutter. The production of discrete side-tones was 
shown by slides and demonstrated by means of a search analyzer. 

(3) The Influence of Room Acoustics upon the Perception of Flutter. 

(4) Method of Measuring Flutter in Sound Tones. 

(5) Flutter in Single- Frequency Tones. (a) The rate of nutter and (b) the 
amplitude of flutter were varied in a 1000-cycle tone to show the accompanying 
changes in the character of the distortion heard. 

(6) Flutter an Important Part of Artistic Vibrato. 

(7) Flutter in Speech and Music. (a) The rate of nutter and (b) the amplitude 
of flutter were varied in excerpts of typical speech and music records to show the 
accompanying changes in the character of the distortion heard. 



REFERENCES 

1 VAN DER POL, B.: "Frequency Modulation," Proc. I. R. E., 18 (July, 1930). 
No. 7, p. 1194. 

CARSON, J. R.: "Notes on the Theory of Modulation," Proc. I. R. E. t 
10 (Feb., 1922), No. 1, p. 57. . 

2 VAN DER POL: loc. cit. 

SHOWER, E. G., AND BIDDULPH, R.: "Differential Pitch Sensitivity of the 
Ear," /. Acoust. Soc. of Amer., 3 (Oct., 1931), No. 2, Part 1, p. 275. 



414 SHEA, MACNAIR, AND SUBRIZI [J. S. M. P. E. 

8 WATSON, G. N.: "Theory of Bessel Functions," Cambridge Univ. Press 
(1922), p. 22. 

4 HANSON, R. L.: "Liveness of Rooms," J. Acoust. Soc. of Amer., 3 (Jan., 
1932), No. 3, p. 318. (Fig. 3 is taken from this paper.) 

6 CURTIS, A. M., SHEA, T. E., AND RUMPEL, C. H.: "The Rapid-Record Os- 
cillograph in Sound Picture Studies," /. Soc. Mot. Pict. Eng., XVHI (Jan., 1932), 
No. 1, p. 39. 

KELLOGG, E. W.: "A New Recorder for Variable- Area Recording," /. Soc. 
Mot. Pict. Eng., XV (Nov., 1930), No. 5, p. 653. 



DISCUSSION 

MR. SHEARER: There is hardly any need for me to emphasize that the care 
and trouble taken in preparing these records was no small task. I dare say it took 
a group of men many weeks, if not months, to prepare the material, and we cer- 
tainly have to thank Mr. MacNair, his associates, and company for producing this 
excellent introduction to the Society of these recording troubles by their true 
names. 

I have always maintained that the tone purity of recording is of more impor- 
tance than the correct frequency characteristic. We have been a little too 
prone to regard good sound as something that merely has the correct frequency 
characteristic. We see how readily extremely small amounts of flutter, three- 
quarters of one per cent, one per cent, three per cent, at various frequencies can 
utterly ruin the sound record; and when we consider that these distortions can 
enter into the original recording, in the printing to some extent and under ex- 
treme circumstances, and also in the reproduction of sound, we can see that a 
sound runs many chances of having this extreme kind of distortion introduced 
into it. If the sound arrives at the theater with less flutter than half of one per 
cent, it really is quite remarkable. 

I believe that flutter trouble is much more noticeable than a small departure 
from a true frequency characteristic. I should like to re-emphasize the fact that 
the various rates of flutter that we heard have a varying effect upon the amount 
of distortion that is apparent. I get the impression that the rates at 96 cps., 
or thereabouts, are more damaging than the rates above that, although, of course, 
the content of the record has something to do with exactly what rate has the 
most effect upon it. 

In dealing with flutter introduced purposely into the record, such as the vibrato 
in singing, we obtain some idea of how critical a singer's vibrato is toward pro- 
ducing a pleasant tone. I believe we notice vibrato errors in singing more than 
any other defect in the quality of their voices. 

MR. KELLOGG: I feel that I am a veteran as to experience in listening to the 
effects of speed variations, or "wows," upon music and speech. My first such 
experience occurred in 1919, when working with Chester Rice on what we hoped 
might be a system of secret telephony, which consisted in distorting speech 
beyond recognition by recording it upon a wire and then taking it off with a 
moving pick-up. Our rate of motion back and forth with the pick-up was 
probably not more than twice per second, so it was a fairly slow "wow," but it 



Nov., 1935] FLUTTER IN SOUND RECORDS 415 

amounted to 25 or 30 per cent. What amazed and at that time discouraged us 
in the work was that, no matter what we did, we could not seem to make speech 
entirely unintelligible. It is the wonderful "toughness" of speech in that respect 
that has made radio and talking movies. 

We called our distorter our "wow- wow" machine, because that term seemed 
to describe the way things sounded that we put through it. So far as I know, 
that was the first use of the term to designate a speed variation. 

A few years later I had a change of heart. Instead of trying to produce 
"wows," I began trying to eliminate them. I worked hard on the problem and 
discussed it at some length at the Washington Convention in 1930. 

This admirable demonstration supplies what I have wanted to hear for years; 
namely, "wows" of known magnitude introduced in order to study the effects 
of various kinds and amounts upon quality. 

There are two or three questions : A vibrato in a single voice is unquestionably 
regarded by many as pleasant. It is provided in musical instruments, such as 
organs, and is produced when desired by violinists, although the amplitude of 
the "wows" so introduced is, I believe, less than that found in many vocal rendi- 
tions. I wonder whether, in the case of a number of tones or voices sounding 
together, it makes any difference whether the "wows" are all in phase, as occurs 
when they are introduced into recorded music by speed fluctuations in a machine, 
or in random relation, as occurs with separate voices or instruments. I wonder 
also, whether the live-room or the dead-room test provides the more reliable indi- 
cation of what is tolerable in "wows" or speed fluctuations. I confess to have 
found it difficult to establish any definite relationship between the fluctuations 
that are noticeable on steady tones in a live room, and those that are found to 
be distinctly objectionable in musical reproduction. In musical reproduction it 
is amazing how much one can stand for, not without some loss of quality, but 
before people begin to complain. Have you made any tests that would indicate 
whether head-phone tests, or dead-room or live-room tests are better guides to 
how much one can stand before the sound begins to sound "sour"? 

MR. MACNAIR: There has been some argument in the Acoustical Society 
recently as to exactly what vibrato was, and I was careful to state that frequency 
modulation was only one part of it. It seems to be well established that a small 
amount of amplitude modulation always accompanies the frequency modulation, 
and just why the 6-cycle modulation sounds best, I do not know. It probably 
has to do with the ease with which that frequency can be produced, because 
with instruments like the violin it is often slower than that. I do not know the 
answer as regards what happens when a chorus of people sing, each one having 
his own modulation. 

As to the question in regard to what one hears in a live or dead room, there are 
perhaps several answers. In the demonstration we tried to show that the fre- 
quency modulation certainly sounded different in a dead room from what it did in 
a live room. Whether the tests should be carried out in a dead room or in live 
room, I believe, depends upon the circumstances. 



A PORTABLE FLUTTER-MEASURING INSTRUMENT* 
R. R. SCOVILLE** 



Summary. A portable flutter-measuring equipment is described which may be 
used to measure the uniformity of motion of recording and reproducing machinery. 
The instrument filters out noise components, magnifies the frequency modulation, con- 
verts it to amplitude modulation, and finally demodulates it to obtain an indication of 
the frequency variation present in the original signal. 

In flutter-correction work adjustments are made on the reproducer under the guidance 
of the measuring set. The cumulative effects of change of pressure or position of the 
guide shoes, alignment of sprockets or bearings, although individually small, are often 
considerable. To study flutter in recording machines 3000-cycle records are made, 
which are subsequently reproduced on known flutter-free machines and measured with 
the instrument. 

Frequency modulation or flutter is caused by non-uniform motion 
in sound recording and reproducing machinery. Until recently the 
practical reduction of flutter in the studio and in the theater has been 
a troublesome problem. This is due to the extremely small amplitude 
of the variations involved and to the difficulty in detecting the me- 
chanical cause of the trouble. But with the development of a port- 
able instrument sufficiently sensitive to measure these complex fre- 
quency variations, a new and much needed technic for correcting 
flutter has been made available for field use. 

Fig. 1 shows the external appearance of the equipment. The in- 
strument is of the direct-reading type, with a range of sensitivity 
between 0.02 and 3.0 per cent. This has been found to be satis- 
factory for all conditions of frequency variation encountered in prac- 
tice. The readings of the instrument represent the per cent change 
above and below the mean frequency being measured, and are inde- 
pendent of the rate of the variation, except for flutter rates below two 
per second, and except also as affected by the optional use of filters 
which aid in thte determination of the flutter rate. 

The equipment consists of two cases weighing about 35 pounds 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** Electrical Research Products, Inc., Los Angeles, Calif. 
416 



FLUTTER-MEASURING INSTRUMENT 417 

each. It is energized from a 110- volt, 50-60 cycle a-c. line, and no 
batteries or additional equipment are needed. For general flutter 
measurement work, the output of a theater reproducing system is 
connected directly to the measuring instrument. When exact de- 
terminations of the flutter frequencies are required, an oscillograph 
may be connected to the instrument for analysis work, but this is 
seldom required. 

The principles underlying the production and measurement of 
frequency modulation, particularly as applied to recording and re- 




FIG. 1. Flutter-measuring instrument set up for 
operation. 

production, have been published elsewhere. 1 The instrument to be 
described herein operates in the following manner: A flutter-free 
3000-cycle record is reproduced on the machine to be measured, and 
the output is connected to the flutter-measuring equipment. The 
instrument circuits first filter out noise and extraneous frequencies 
which would otherwise interfere in the measurement. The wave 
is then heterodyned by a variable-frequency oscillator in such man- 
ner that a difference-frequency of predetermined value is obtained. 
Consequently the mean input frequency need be only approximately 
3000 cycles. The resultant wave of definite mean frequency and 
amplitude is then impressed upon a frequency-discriminating network 



418 R. R. SCOVTLLB [J. S. M. P. E. 

which converts the frequency-modulated wave to one having ampli- 
tude modulation. The transformation is such that the percentage 
of amplitude modulation obtained is proportional to the percentage 
of frequency modulation and independent of the rate of this modulation. 

After further amplification, the amplitude-modulated wave is im- 
pressed upon a demodulator, which recovers the envelope frequency 
and measures it. Since the envelope amplitude is related to the per- 
centage of frequency variation and is of the same rate as that of the 
original speed variation, it is made to operate a meter calibrated di- 
rectly in percentage of frequency variation. Headphones or an os- 
cillograph may be plugged into the instrument to determine what 
flutter rates are present. A segregation of the flutter frequencies 
present is made possible by the use of a filter incorporated in the in- 
strument, by means of which readings may be made of either the low- 
frequency (1 to 20 cps.) or the high-frequency components (20 to 
130 cps.). These respective bands may also be connected to an ex- 
ternal oscillograph for more detailed analysis, if desired. 

Despite the intricate transformations performed in the flutter- 
measuring instrument, the manual operations required to obtain a 
reading are simple and logical. The signal is adjusted to a designated 
amplitude, and the calibrating dial is manipulated to compensate for 
deviations of the mean input frequency, until the meter directly above 
the dial (Fig. 1) indicates the designated value. After setting the 
meter-scale dial (lower left, Fig. 1) to the correct position, depending 
upon the amount of flutter in the source, the percentage of frequency 
variation is indicated upon the meter shown in the lower case of the 
instrument (Fig. 1). The meter-scale dial provides for full-scale 
meter indication for percentage of frequency variations of ==0.1, 
0.3, 1.0, and 3.0 percent. Having determined the total per- 
centage of frequency variation in the source, the classification of the 
variation as to rate is made by throwing over the low-pass filter key 
where flutter components having rates up to 20 cps. are measured. 
Similar measurements are made in the high-range position, of com- 
ponents from 20 to 130 cps. This feature is of special value in locat- 
ing sources of flutter for purposes of naking adjustments. 

Approximate determination of the rate of variation of the flutter 
pattern is made by listening with telephone receivers introduced at 
appropriate points in the circuit. If the rate is low, the flutter may 
be readily distinguished and counted in reference to time; whereas, if 
it is high, a note of the corresponding frequency may be heard by 



Nov., 1935] 



FLUTTER-MEASURING INSTRUMENT 



419 



connecting into the demodulator circuit. Either a cathode ray or 
an oscillograph of the recording type may be plugged in, if desired, 
to analyze the flutter pattern. Figs. 2 and 3 are oscillograms of the 
output of the flutter-measuring instrument. Fig. 2 was made using 
the low-pass filter, which excluded flutter frequencies greater than 
about 25 per second. A frequency of about 20 per second is very 
noticeable, together with a component having a rate between 2 and 
3 per second. The 20-cps. rate was found to correspond to the rota- 
tional speed of one of the drive shafts, while the second was the 
result of unbalance of the flywheel. Fig. 3 shows the flutter compo- 
nents present in the higher range. In this case the dominant fre- 



Vioo Sec. 




FIG. 2. Low-frequency flutter pattern, with a prominent 20-per second 
component, and an additional variation occurring 2 -3 times per second. 



V Sec. 




FIG. 3. 



High-frequency flutter pattern showing the 25-150 per second com- 
ponents. Note the prominent 96 per second variation. 



quency is found to be 96 cps. which is the rate at which the sprocket 
teeth engage the film. 

In flutter correction work, adjustments are made on the reproducer 
under the guidance of the flutter-measuring instrument. Fig. 4 
shows the instrument set up in a projection booth and adjustments 
being made on the machine. The procedure varies with the nature 
of the equipment involved; but, in general, adjustments are made 
while the machine is operating by observing the indications of the 
flutter-measuring instrument. The application of finger pressure at 
certain points in the film path may reduce the flutter and indicate to 
the operator the need of certain mechanical adjustments. Change 
of sprockets often produces a considerable improvement. The rela- 
tive alignment or eccentricity of sprockets or rollers, defective gears, 
and unbalance of the flywheel system, may all be contributing flutter 
factors. The measuring equipment, by informing the operator of 



420 R. R. SCOVILLE [J. S. M. p. E. 

small improvements made during the adjustments, each of which 
might by itself be inaudible, makes possible a cumulative improve- 
ment that may be of very observable magnitude. 
In a recent case of flutter correction, measurement of the machine 




FIG. 4. Operation of flutter-measuring 
instrument in the projection room of a 
theater. 

disclosed a frequency variation of approximately 96 cps., due to 
sprocket hole pull, of 0.9 per cent and a low-frequency reading of 
=*=0.25 per cent, mainly at six cps. The rates mentioned were 
identified by a head-phone test, but could have been more precisely 
analyzed by an oscillograph. The pressure of the guide shoes in the 
vicinity of the sound sprocket was changed, and the reading dropped 
to 0.7 per cent. The sound sprocket was removed, inspected, and 



Nov., 1935] 



FLUTTER-MEASURING INSTRUMENT 



421 



found to be undercut due to wear. Replacement resulted in a reduc- 
tion to ==0.5 per cent, or approximately half the original value. Al- 
though this was not regarded as entirely satisfactory, attention was 
shifted to the low-frequency variation. Measurement showed the 
sound sprocket to have excessive eccentricity. This was corrected. 
The low-frequency reading dropped from 0.25 to 0.18. Similar 
treatment of a second sprocket near the filtered drive decreased the 
reading further to 0.15 per cent. Next, the alignment of all sprock- 
ets and rollers adjacent to the scanning point was carefully checked, 




FIG. 5. Reproducer for flutter measurements of test recordings. 

and changes made as required. The reading then dropped to 0.12, 
which was one-half the original amount. Switching over to the high- 
frequency measuring range disclosed a reduced reading, and after 
further adjustment of the guide shoe the indication dropped to 0.32 
per cent, or about one-third the original variation. It was noticed, 
however, that slight finger pressure upon one side of the film caused 
the reading to drop appreciably. The guide shoe assembly was re- 
moved and provision made to shift the angle of the guide shoe axis 
relative to the film. On reinstalling and adjusting to the optimum 
position, the reading on the high-frequency range was found to have 
dropped from the original value of 0.9 per cent to 0.25 per cent, 



422 R. R. SCOVILLE 

which was regarded as satisfactory. After some further work on the 
machine, the low-frequency reading came down to a minimum of 
=*=(X08 per cent, which was approximately one-third the original 
value. Listening tests confirmed the over-all improvements shown 
by the measurements, although each individual improvement was too 
small to be noticeable. 

In extensive field tests recently conducted, the average high-fre- 
quency type of flutter was reduced to half its original value by means 
of the technic herein described. At the same time the average low- 
frequency flutter was reduced to two-thirds. None of the reductions 
were of such a nature that the unaided ear could have effected the 
improvement, which in the aggregate was considerable. 

In flutter correction work upon recording machines, the procedure 
is somewhat different from that just outlined. Several test record- 
ings are made and subsequently reproduced by a relatively flutter- 
free reproducer, so that the measured flutter is substantially that re- 
corded upon the film, rather than that introduced by the reproducer. 
Measurements are made with each sample and the relative readings 
are compared. Fig. 5 shows a practically flutter-free reproducer 
used in measuring film recordings. It consists of a drum mounted 
upon the turntable of a high-quality disk recording machine, and a 
scanning means for reproducing film wound around the drum. This 
method, originally proposed and used by Bell Telephone Labora- 
tories, has given very satisfactory results. At present, the flutter 
produced in film recording machines is considerably less than that 
found in average projection machines; consequently the measure- 
ments of recorded flutter are more delicate than those for film repro- 
duction. 

Marked improvements in sound quality almost invariably follow 
the intelligent application of this technic. The flutter-measuring 
instrument, by indicating the approximate nature of the frequency 
variations present, suggests the method of correction. Practical re- 
sults have shown that some form of measuring instrument is an al- 
most indispensable tool for correcting flutter. 

REFERENCE 

1 SHEA, T. E., MACNAIR, W. A., AND SUBRIZI, V.: "Flutter in Sound Records," 
J. Soc. Mot. Pict. Eng., XXV (Nov., 1935), No. 5, p. 403. 



LIGHTING FOR TECHNICOLOR MOTION PICTURES* 
C. W. HANDLEY** 

Summary. Some of the studio illuminating equipment used for lighting in 
Technicolor productions is described, together with brief discussions of the spectral 
characteristics of the various units. Included are the broadside, supplementary light- 
ing units, the Sunlight arc, the Sun Spot and Solar Spot, diffusing mediums and color 
filters, and arc silencing devices and methods. 

The introduction of the Technicolor three-color process of color 
photography into the motion picture studios brought forth the need 
for improved studio lighting equipment. All known color processes 
involve filtering or breaking up the light entering the camera into the 
primary colors, and recording each color upon a separate emulsion. 
Obviously, in a three-color process, a lower intensity of illumination 
falls upon each of the three negatives than upon the single negative 
used in black-and-white photography, because the latter receives all 
the rays passing through the camera lens. It is therefore necessary 
that the stage illumination be of higher intensity than required for 
black-and-white photography in order to attain corresponding photo- 
graphic speed. Furthermore, since the color-sensitivity of the three 
negatives and the color-balance of the process as a whole is designed to 
render accurate color tones under daylight illumination, it is highly 
desirable that the light-source used in the studio should have a spec- 
tral energy distribution conforming closely to that of natural sun- 
light. 

It is true that the same photographic effect can be attained with a 
light-source differing in quality from that of sunlight by using suitable 
filters. This, however, involves the absorption of a portion of the 
light reflected from the set, and a consequent reduction of photo- 
graphic speed. The fact must always be recognized that filters 
never add light of any color. They merely reduce the intensity of 
the colors they are designed to suppress. Any increase of tempera- 
ture on the set over that experienced in black-and-white productions 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** National Carbon Co., Los Angeles, Calif. 

423 



424 



C. W. HANDLEY 



[J. S. M. P. E. 



is highly undesirable. It is accordingly evident that changes in 
lighting equipment must be of a character to provide an increase of 
photographic light of daylight quality without exceeding the amount 

of radiant heat projected upon 
the stage by the lighting equip- 
ment used for black-and-white 
photography. 

THE BROADSIDE 

The white-flame carbon arc 
has long been recognized as a 
light-source of high photographic 
efficiency and as one providing 
photographic effects essentially 
equivalent to those of sunlight. 
The old type of broadside lamp, 
burning l /z by 12 -inch white- 
flame carbons at 40 or 45 am- 
peres were extensively used in 
the days of silent pictures. The 
mechanism of these lamps is of 
such design, however, that it is 
practically impossible to adjust it 
so that it will operate with the 
quietness necessary when used 
in proximity to sound recording 
equipment. Furthermore, higher 
intensities of illumination than 
these units are capable of supply- 
ing are required to meet the 
needs of the Technicolor process satisfactorily with a reasonable 
number of lamps upon the set. Since the broadside, used for floor 
lighting, and the scoop, for overhead lighting, provide the broad level 
of general illumination for the set as a whole, they constitute the 
most important elements of the lighting equipment. It is these units 
that establish the general color-tone of the scene in Technicolor 
photography. 

The research laboratory of the National Carbon Company under- 
took the development of a new carbon to fulfill the specific needs of 
this new photographic process. The result of the research was a 




FIG. 1. MR-29 twin-arc broadside 
lamp. 



Nov., 1935] LIGHTING FOR TECHNICOLOR PICTURES 425 

metal-coated carbon 8 millimeters in diameter, designed for opera- 
tion at 40 amperes. Its characteristics have been described in detail 
by Joy, Bowditch, and Downes. 1 Due, in part, to the high current- 
density at which these carbons are operated, their light departs some- 
what from the normal characteristics of the white-flame carbon arc 
and takes on more of the character of the high-intensity arc. In fact, 
through the photographically effective range, the relative intensity of 




FIG. 2. MR-27 scoop. 

radiation at various wavelengths is almost indentical to that of the 
13.6-mm. high-intensity projector carbon arc operated at 125 amperes. 
The development of a suitable lamp for use with these carbons was 
then accomplished through cooperation with an established lamp 
manufacturer. Two lamps were designed and made available to the 
motion picture studio : the twin-arc broadside lamp, MR-29, shown 
in Fig. 1, and the twin-arc scoop, MR-27, shown in Fig. 2. A detailed 
discussion of the development of these lamps has been given by Mole 2 
together with a statement of the specifications that had to be fulfilled 
in adapting them to color photography. These units burning the new 



426 



C. W. HANDLEY 



[J. S. M. P. E. 



National motion picture studio carbons deliver somewhat more than 
the required 200 foot-candles at 15 feet; give an even distribution of 
light, constant in quality and intensity; provide a spectral energy 
distribution very similar to that of sunlight; and fully fulfill the re- 
quirements of silence imposed by the sound technicians. They have 
proved highly efficient for black-and-white photography, and prac- 
tically a necessity for color photography. A comparison of the spec- 
tral energy distribution of the light from these lamps with that of 



ULTRA-VIOLET VIOLET BLUE GREEN YELLOWORAlfcE RED 



STUDIO CARBON ARC 
4OAMPS.-37.5VOLTS 




3000 



7000 



4000 500O 6OOO 

WAVE LENGTH IN AMGSTROM UNITS 
FIG. 3. Spectral energy distribution of studio carbon arc and sunlight. 

natural sunlight is given in Fig. 3. A distinct advantage of this light, 
as pointed out by Joy, Bowditch, and Downes, 1 is that more than 40 
per cent of the radiant energy emitted is photographically effective. 

SUPPLEMENTARY LIGHTING UNITS 

While broadsides and scoops provide the general level of set il- 
lumination and are highly satisfactory for the front and side lighting, 
their use without supplementary equipment would result in flat and 
uninteresting photographic effects. Accordingly, very powerful 
lighting units are placed in elevated and other strategic positions when 
strong shafts of light are required to pierce the even intensity of il- 
lumination supplied by the broadsides and the scoops. These larger 
units are used also for increasing the intensity of light in any given 
area, thereby separating points of special interest from the remainder 
of the set. The units used for this supplementary lighting are the 
spotlights and sun arcs, powerful carbon arc lamps that utilize the 
high-intensity principle first applied to searchlights and later exten- 
sively adopted for motion picture projection, 



Nov., 1935] LIGHTING FOR TECHNICOLOR PICTURES 427 

THE 80-AMPERE ROTARY ARC SPOTLIGHT 

The 80-ampere rotary arc spotlight is used for back -lighting and to 
increase the intensity of illumination at any point where projected 
light is required, where the increase desired does not demand the 
power of a sun arc. It is in regular use in most of the studios and has 
been adapted to sound by the use of fiber gears which reduce mechani- 
cal noise. Some of them are fitted with snap-switches to cut out the 
control motor when the unit is close to the microphone. They are 
operated at 75-80 amperes with 50-55 volts at the arc. The operat- 
ing element of an 80-ampere rotary spotlight is shown in Fig. 4. 




FIG. 4. Element of 80-ampere rotary arc spotlight. 

To insure full efficiency and uniform photographic effect, reasonable 
attention should be given to the maintenance of these lamps and to 
preserving correct conditions of operation. A recent examination of 
several of them in operation revealed considerable variation in the 
quality and quantity of light emitted. This resulted from various 
causes, all easily removed or prevented. A badly pitted and soiled 
condenser was found to cut out as much as 40 per cent of the total 
light output. A variation of l /z inch in the arc-gap (not uncommon 
in practice) changed the spectral energy distribution sufficiently to 
be noticeable in the Technicolor negative. In order to arrive at a 
standard for this type of unit in one instance, a lamp was fitted with a 
clean condenser free from pit marks, the best arc length (approxi- 
mately x /2 inch) was maintained, a spot three feet in diameter was 
focused upon a white wall thirty feet from the unit, and the direct 
light was measured with a standard Weston photometer fitted with a 
filter that reduced the intensity to within the limits of the instrument. 



428 



C. W. HANDLEY 



[J. S. M. P. E. 



The figures so obtained are used in checking other units of this type. 
The spectral energy distribution of the 80-ampere rotary spotlight 
is higher at the blue end of the spectrum than that of the broadside 
lamp. A satisfactory color balance is maintained by the use of 
straw-colored gelatin in front of the condenser. Although this filter 

is very light, it reduces by more 
than 20 per cent the photo- 
graphically effective radiation of 
the lamp. The need for the 
development of a new lamp, to 
give the desired spectral energy 
distribution without the use of a 
filter, is therefore indicated. In 
all probability such a lamp should 
have a different carbon trim from 
that now being used in rotary 
spotlights. 

THE SUNLIGHT ARC 

Two sizes of Sunlight arc are in 
common use, the mirror diameters 
of which are, respectively, 24 
inches and 36 inches. They are 
designated as 24-inch Sun arcs 
and 36-inch Sun arcs. A 36-inch 
Sun arc is shown in Fig. 5. These 
lamps are used where the highest 
intensity of projected light is re- 
quired, as in back-lighting when 
the action calls for a high level 
of foreground illumination; where 
well-defined shadows are desired; 
where a clearly defined streak of 
light cuts through the general illumination; and for producing high 
intensities of general illumination of great penetration. In the latter 
case diverging doors composed of strips of cylindrical lenses are often 
used in front of the lamp houses. 

The spectral energy distribution of the light from this lamp is 
similar to that from the rotary spotlight, and, likewise, requires the 
use of a light straw filter to establish accurate color balance. There is 




FIG. 5. 36-inch Sun arc. 



Nov., 1935] 



LIGHTING FOR TECHNICOLOR PICTURES 



429 




bp .d bo bo 

I bo fj bo d bo 

'3 ^ '-is .S ' .2 

j d *rt .d -M d -M 

bo oJ W),d bo .S 
P UH a5 vd bo U2 bo 

^ JH ji 4J ^4 ^3 

"! a s a 

<o rn </* 



.19 I -S 



S'l 



M 



10 



.2 



IB 

O O5 

* X 



W o 
w X 



S W2 



X 



S S 



-4-> OJ *< eo 

^Si s 

I-H en 00 
V Sf<N < 

x d ^ " 

5 'g X 

6 S 
06 > 



il 

IS 



Si 



r X 



a 

S -a 



N di 



a 



S aT 

? -5 

CO ^~ 



^ s 

S o 

| e 
I 



S 3 ! 

" ^) 
rt ^ 

!'" 



00 



"8 



s 

I 

a 



1 

O O 










03 .2 

d 

03 *""> 



430 C. W. HANDLE Y [J. S. M. P. E. 

accordingly evident need for further development of this type of unit 
to eliminate the necessity of using a filter. 

THE 36-INCH SUN SPOT 

The 36-inch Sun spot is similar in appearance and design to the 36- 
inch Sun arc except that it uses as a light-source a 10-kw. special 
tungsten lamp. It is used where a color contrast is desired and where 
warmer tones predominate; as behind windows, where the effect of 
the increased red radiation creates the illusion of the yellow-orange of 
afternoon sunlight. 

JUNIOR SOLAR SPOT 

The Junior Solar spot is a newly developed unit fitted with a special 
prismatic front lens and a spherical mirror. It is equipped with 
a 2000-watt, G-48 tungsten lamp which may be moved to a flood 
position or to a focus where it delivers a highly concentrated beam. 
The type of trim, as well as the arc current and voltage used by the 
different types of lighting units, is shown in Table I. 

DIFFUSING MEDIUMS AND LAMP COLOR FILTERS 

As previously stated, some of the high-intensity Sun arcs and 80- 
ampere, rotary-arc spotlights are fitted with straw-colored gelatin 
filters to cut out an excessive amount of blue. In addition to these, 
gelatin hangers of various colors are available so that the spectral 
energy distribution of any lamp may be changed at will to suit the 
requirements of the scene. Frosted gelatin hangers are used to soften 
the light from certain lamps. The Sun arcs have, as auxiliary 
equipment, diverging doors for use in spreading the beam on its 
horizontal transverse axis. The broadsides are fitted with " Factor- 
lite" glass screens. These are sand-blasted on one side and molded 
on the other, making excellent diffusing mediums. 

ARC-SILENCING DEVICES AND METHODS 

Methods of silencing arc lights have been perfected by the studios. 
In 1930, W. Quinlan, Chief Engineer of the Fox Studios, produced and 
equipped that studio with complete sound filtering devices. These 
units consist of high-capacity electrolytic condensers connected 
across the bus-bars of the generator, and individual choke-coils 
mounted as an integral part of each lamp. 3 Other studios use high- 
capacity condensers of the dry type, and individual choke-coils for 



Nov., 1935] LIGHTING FOR TECHNICOLOR PICTURES 431 

the various types of units. L. Kolb, Chief Electrical Engineer of 
M-G-M Studios, has developed larger choke-coils rated at 1000 am- 
peres, which are mounted upon rollers and may be used at the power 
house. These developments, together with the previously mentioned 
development of the new motion picture studio carbon arc, make 
possible the use of 100 per cent carbon arc illumination of the motion 
picture stage without the slightest interference with sound recording. 

REFERENCES 

1 JOY, D. B., BOWDITCH, F. T., AND DOWNES, A. C.: "A New White-Flame 
Carbon for Photographic Light," /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), 
No. 1, p. 58. 

2 MOLE, P.: "A New Development in Carbon Arc Lighting," J. Soc. Mot. 
Pict. Eng. t XXII (Jan., 1934), No. 1, p. 51. 

3 Report No. 2, Producers and Technicians Committee, Acad. Mot. Pict. Arts & 
Sciences (May 7, 1930). 



ERRATUM 

In the paper entitled "The Technical Aspects of the High-Fidelity Repro- 
ducer," by E. D. Cook, published in the October, 1935, issue, equation (10) on 
p. 304 should read 

/t = juoU + 00-1 



REPORT OF THE STUDIO LIGHTING COMMITTEE* 

Several years ago a comprehensive program of making available 
to the industry complete information on illuminants, equipments, 
and practices in studio lighting was planned. Owing to the extent of 
the subject, it was deemed advisable to present this material in four 
reports, each covering a particular phase of studio lighting. 

The first report was to deal with illuminants, their characteristics, 
available sizes, etc.; the second to discuss the various lighting equip- 
ments; the third, power supplies and distribution methods; and the 
fourth, lighting practices. To carry out this program it was neces- 
sary that the Committee be kept intact through two administrations 
of the Society, and it was the thought that all this material might 
eventually be published as a single booklet, a sort of handbook on 
studio lighting. 

The first two of the reports were presented before the Society and 
published in the JOURNAL.* 2 Unfortunately, because of changes of 
the Committee personnel, reports covering the third and fourth 
parts of the program were never presented. The present Committee, 
which includes in its personnel four of the original Committee that 
formulated the program, believed it a good plan to bring sections one 
and two up to date, as well as to present the third part covering studio 
power supplies and distribution methods. 

(1) IMPROVEMENTS IN LIGHT-SOURCES 

Arcs. One of the most outstanding arc developments for studio 
lighting since the presentation of the original report has been the 
high-intensity, white-flame carbons. These carbons, 8 millimeters in 
diameter, are designed for use in general lighting equipments such 
as broadsides and scoops. They are copper-coated, and operated at 
35-40 amperes, or at two and one-half times the current density at 
which the older Y2-mch carbons were operated, thus giving approxi- 
mately 25 per cent more light for the same wattage. More complete 
data on these carbons has been given by Joy, Bowditch, and Downes. 3 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
432 



STUDIO LIGHTING COMMITTEE 433 

Further improvements in arc carbons for studio lighting have been 
in the nature of refinements. A l / 2 X 12-inch cored positive carbon 
for the 80-ampere rotary has been made available, with better per- 
formance features. A 5 /i 6 -inch diameter copper-coated negative 
carbon is now recommended in place of the 3 / 8 -inch negative; also, 
the 3 / 8 -inch negative replaces the 7 /ie-inch negative. 

Mazda Lamps. The original report described characteristic incan- 
descent lamps operating at filament temperatures up to about 
3100K. Since that time, lamps have been made available with fila- 
ment temperatures up to 3450 K. This is interesting in view of the 
fact that tungsten melts at 3655. The advantage of this higher 
temperature is that the blue-violet radiation is increased 270 per cent, 
whereas the red radiation is increased 155 per cent. In other words, 
the lamps known as the Movieflood and the Photoflood give nearly 
twice as much blue- violet light for the same amount of red, and at 
increased efficiency. They are particularly advantageous for color 
photography. Complete data on these lamps were given in the 
JOURNAL by Farnham. 4 

A new bipost has replaced the mogul screw base on the 2000-watt 
G48 bulb lamp and the prong base of the 5000-watt G64 bulb lamp. 
This new base is much more rugged and positions the light-source 
with greater precision in the optical axis of the projectors. 

A group of incandescent lamps ranging in wattage from 1000 to 
5000 watts and capable of burning base upward, have been made 
available. These lamps employ the bipost base, and are particularly 
adapted for use in the new elliptical reflector spots. Their long tubu- 
lar bulbs, with the light-source close to the end, result in minimum 
shadowing of the reflector by the bulb, and the base-up burning fea- 
ture causes the bulb-blackening to occur outside the limits of the 
reflector (Fig. 1). 

There has been considerable development in gaseous conductor 
sources, and a sodium vapor, as well as a high-intensity mercury 
lamp, has been made commercially available. However, both these 
sources possess both light and performance characteristics that render 
them unsuitable for motion picture studio lighting in their present 
form. 5 

(2) DEVELOPMENTS IN LIGHTING EQUIPMENT 

The second report, which discussed studio lighting equipment, de- 
scribed a number of units for both general and modeling lighting, 



434 



STUDIO LIGHTING COMMITTEE 



[J. S. M. P. E. 



practically all of which are in general use today. There has been a 
very definite trend away from the one- and two-lamp broadsides, and 
toward the use of the "rifle" unit, because of the latter 's much higher 
efficiency. Overhead suspension arrangements have been developed 
to permit the rifle reflector to be hung by its yoke above the set, thus 
gaining greater illumination intensities than were possible from the 
trough reflectors primarily designed for this service. 

Contemporarily with the development of the 8-mm., high-intensity, 
white-flame carbon was the designing of an improved twin-arc broad- 
side lamp to take full advantage of the new 
carbons. The particular feature of this lamp 
is the addition of a voltage-operated coil con- 
nected across the arc, whose flux opposes that 
of the current coil. This results in a very 
much steadier lamp, with constant electrode 
spacing. Further refinements include the use 
of a chromium reflector and ball bearings in 
the carbon feed-mechanism, as well as a more 
rugged and compact type of ballast resistance. 
A sand-blasted glass diffusing screen is used 
with this lamp in place of the old inefficient 
"silks." This unit is also available in the 
"scoop" form, as well. Additional data were 
given in a paper by P. Mole. 6 

An important development in incandescent 
lighting equipment has been the elliptical re- 
flector spotlamp. This unit makes use of 
the ability of an elliptical reflector to pick up 
a large solid angle of light-flux from a source 
at one focal point and concentrate it at the other focus. About half 
the complete ellipsoid is used, and an iris diaphragm is placed just 
in front of the opening of the objective lens of large diameter, which 
is placed just beyond the point at which the light-rays cross, and 
images the diaphragm upon the area illuminated. The spot of light 
produced is uniform and sharply defined, and is adjustable in 
diameter. Four adjustable straight-edged shutters near the iris 
make spots of a variety of shapes and sizes possible. The gain in 
illumination over the conventional lens spot is about 200 per cent 
both at the narrow and wide beam spreads 7 (Fig. 2). 
There is now available a small incandescent unit called the "Handi- 




FIG. 1. Base-up 
burning incandescent 
lamp, used in certain 
types of reflector 
spots. 



Nov., 1935] 



STUDIO LIGHTING COMMITTEE 



435 



lamp," which usually uses the 1000- watt tubular bulb lamp, either 
clear or frosted. It combines compactness and light weight, and is 
frequently mounted upon or close to the camera blimp to provide a 




(U M- 

>* 

55^ 






o' 



high intensity of illumination over a localized area, generally in 
"close-ups." A fluted chromium- or even silver-plated reflector, 
about 10 inches in diameter, is used. The flutes in the reflector 



436 STUDIO LIGHTING COMMITTEE [j. s. M. P. E. 

spread the beam more in one direction than in the other, and the 
amount of spread is controlled by moving the lamp in and out of the 
reflector (Fig. 3). 

The No. 4 photoflood lamp developed originally for still photog- 
raphy, has found its way into the motion picture studios, and has 
prompted the designing of an unusually efficient reflector unit, shown 
in Fig. 4. The diffuse-finish aluminum reflector combines high re- 
flection efficiency with a large light pick-up. This equipment is 
especially advantageous for use upon location or with expeditions 
where weight and size of equipment, as well as available power supply, 
are important factors. One of these units consuming 1000 watts is 
capable of giving illumination intensity equal to that furnished by 
2500 watts in the usual types of lamps. 

A new development in lens type spots that has great promise is 
the so-called Junior spotlamp. This unit substitutes the conventional 
plano-convex condenser with a Fresnel lens. Thus, it is possible to 
operate the light-source much closer to the condenser and increase 
the light intercepted by the lens several-fold. A spot employing the 
2000-watt G48 bulb lamp has recently been marketed, and equip- 
ments for both the 5000-watt lamp and the high-intensity arc are 
under development. 8 

The general adoption of supersensitive panchromatic film brought 
with it an unexpected, though interesting, problem. The great 
sensitivity of this emulsion makes it particularly susceptible to re- 
cording irregularities of illumination at low intensities, with the result 
that "spill" light from the equipment mounted overhead and shining 
upon the upper walls of the sets appears quite conspicuously in the 
picture. This has led to the general use of spillshields on the 18- and 
24-inch sunspots. 

"Skylight," or "Pan," is another unit that has grown up from a 
make-shift. It consists of a shallow diffuse reflector about 24 inches 
in diameter, provided with a socket for the 5000-watt lamp and a 
suitable connecting cable. It is usually suspended high upon the 
set, to provide uniform illumination of sky backings or cycloramas. 
The contour of the reflector and its finish are such as to afford a 
fairly uniform light distribution through a wide angle, thus permitting 
the unit to be placed relatively close to the surface being lighted. 

The advent of motion picture photography in color has brought 
with it the necessity of frequently controlling the color of the light 
to produce certain effects. At present this is being done with colored 



Nov., 1935] STUDIO LIGHTING COMMITTEE 437 

gelatins. Gelatins have the disadvantage of fading, and it would be 
desirable to have many of the more frequently used colors as glass 
niters. The glass must either have a wire mesh embedded in it, or 
be used behind a wire screen to prevent its falling to pieces in event 
of breakage. 

To obtain the practically perfect white light necessary for the Tech- 
nicolor three-color process, there has been produced by the Corning 
Glass Works a suitable glass filter known as No. 570, which, when used 
with incandescent lamps operating at 33 lumens per watt, results in 
a light, the color of which meets the requirements of the process 
satisfactorily. 

The original equipment report called attention to hum filters in 




FIG. 5. Remotely operated switchboard, affording greater flexibility 
of lighting control. 

capacities up to 300 amperes; recently, filters of 1000-ampere rating 
have been manufactured. Greater use is being made of portable, 
remote controlled switchboards. These afford greater flexibility in 
operating the lighting equipment and make possible many effects 
not otherwise easy to produce. Fig. 5. shows one type, capable of 
controlling separately, or by master-control, four circuits of 400 
amperes each. A length of 100 feet of control cable terminating in a 
five-gang switch is shown on top. 

(3) POWER SUPPLY AND DISTRIBUTION METHODS 

Practically all the studios making theatrical pictures employ di- 
rect current obtained from motor-generators. It is interesting to 



438 STUDIO LIGHTING COMMITTEE [ j. s. M. P. E. 

note, however, that extensions in the power facilities of several studios 
have been made by the use of transformer banks supplying 125-250 
volts a-c. direct to the lighting equipment. One or two of the eastern 
studios located on d-c. central station distribution lines also use this 
source to supplement the power obtained from their own substa- 
tions. A few very small studios also on d-c. systems obtain their 
power solely from the central stations. 

This almost universal use of direct current results from the fact that 
the d-c. arc is more efficient, quieter, and better adapted to light-pro- 
jecting equipment than the a-c. arc, and nearly all the present studio 
power installations were made at the time the arc source was in more 
general use. If called upon to re-equip the studios completely, many 
electrical staffs would be in favor of employing alternating current 
even to the point of placing transformer banks in each stage, because 
incandescent lamps operate equally well with either direct or alter- 
nating current. The initial investment in copper and rotating ma- 
chinery, as well as the operating and maintenance costs of such a 
power supply, would be very much less. Small or medium-power 
portable motor-generator sets would be brought to the stage when 
necessary to employ arc equipment. 

With the advent of sound recording, direct current was at one time 
regarded necessary, since it was feared that trouble might be experi- 
enced from hum pick-up into the sound channels. Adequate shield- 
ing of both the wiring and equipment has, however, removed the 
possibility of trouble from this source. 

Permanently installed generators for studio service are of 125-250 
volts' rating and from 300-500-kilowatt capacity. They are usually 
flat-compounded, but in a few instances over-compounded. Both 
induction and synchronous motor drives are used. The more recent 
installations employ synchronous motors. It is general to use a 
single motor with a d-c. generator at each end of the shaft and upon 
the same base. Hand-regulation of the d-c. voltage is universal, 
and an effort is made to provide 120 volts at the sets. 

Portable generators mounted either upon trucks or trailers are 
used to supply power when on location, or are frequently used to 
supplement the regular substation on very large sets or in case of a 
heavy shooting schedule. The generators are of the 3- wire type, 
rated at 125-250 volts, 100-300 kw., where motor drive is used, and 
25-200-kw. capacity for gas-engine drive. The motors for portable 
motor-generator sets used by the West Coast producers are wound so- 



Nov., 1935] STUDIO LIGHTING COMMITTEE 439 

as to be capable of operating on either 2200 or 4400 volts, three- 
phase, by suitable change of connection. They can generally be used 
on either 50 or 60 cycles. 

The gasoline-engine driven portable generator outfits are, of course, 
necessary when on location where high- voltage lines are not available. 
Oftentimes small gas-engine generator sets of 25-kw. capacity or even 
less are taken about on the lot to operate a few "booster" lights, this 
practice being simpler than endeavoring to run wires from the studio 
substation or "high line." 

Owing to the pronounced change of speed with load, an inherent 
characteristic of the gas-engine, trouble has been experienced because 
of sudden rises in the generator voltage to 140-150 volts when part of 
the load is suddenly switched off, and before the governor controlling 
the engine throttle valve has had time to operate. This results in a 
sudden change in the light output of the remaining lamps or may cause 
some to fail prematurely. To overcome this difficulty, a quick-acting 
voltage-operated regulator has had to be developed for holding the 
generator voltage in check until the throttle valve operates. The 
necessity for absolute quietness in connection with sound picture 
photography has brought additional problems in the design of the 
"gas wagon." Not only are special mufflers necessary but the entire 
engine and generator must be enclosed in a sound-proof housing. 
A very complete description of several gas engine-generator sets has 
been given by Mole. 9 One studio has made up a portable transformer 
wagon, analogous in its application to the portable motor-generator 
set. The transformers are so wired as to receive 2200 or 4400 volts 
on the primary side, and deliver 120-240 volts at the secondary. 

DISTRIBUTION 

It is quite general practice in the studio substations to employ a 
double bus system of 125-250 volts. Double-throw switches permit 
any or all of the generators to be connected to either bus. In a simi- 
lar manner, the circuits to any of the stages may be connected to either 
of the buses. Such an arrangement has the advantage of isolating 
part of the system in case of trouble. It also makes possible differ- 
ent voltages at various locations, as, for example, a higher voltage for 
those stages at a greater distance, where the line drop is greater. 

Three-wire, 120-240-volt feeders are carried from the main switch- 
board to the several distribution points in underground conduit. 
Within the stages either of two wiring arrangements may be em- 



440 



STUDIO LIGHTING COMMITTEE 



[ J. s. M. p. E. 



ployed : One, regarded as the newer type and shown schematically in 
Fig. 6, brings the feeders into the upper part of the stages, where they 
connect to several remotely controlled switchboards. Four-hole 
plugging boxes are usually permanently connected, each to a switch 
on this overhead switchboard, by a 100-foot length of 3-wire cable. 
When not in use, the plugging box and its cable are fastened overhead. 
When put into use the box is lowered to either the parallel or the stage 
floor and the lighting units plugged in. These switchboards also 
have short copper bus-bars connected directly to the line and to 
which No. 2, 3-wire extension cables may be connected. These 
cables are, in turn, connected to spiders or coupling blocks. Heavy- 
current devices, such as sun arcs and rotaries, are then attached to the 




FIG. 6. Schematic diagram for permanently installed, 
remotely controlled overhead switchboards. 

coupling blocks. The switches of these remotely controlled boards 
are operated by small tumbler switches at the end of a 100-foot 
flexible conductor, which may be placed at any part of the set. 

The second and older method of power distribution on the stages, 
outlined in Fig. 7, consists in placing a number of outlets along the 
walls of the stages near the floor, to which the feeders from the sub- 
station are attached. These outlets comprise a large-capacity switch 
known as the "bull" switch, terminating in three heavy, short bus- 
bars , 0000 feeders being carried from this point to the portable switch- 
board. This may be operated by an attendant at the board or 
remotely controlled. The plugging boxes for the various lighting 
units are usually attached directly to this board, although in some 
instances, when greater length of cable is required, an extension may 
be introduced terminating in a splicing block, to which the plugging 



Nov., 1935] 



STUDIO LIGHTING COMMITTEE 



441 



box cable is attached. The heavy-current units, such as the sun 
arcs or rotaries, are also clamped to the spiders or splicing blocks. 
The spiders to which these heavy-current devices are attached are 
kept "hot"; that is, their extension cables by-pass the switches on 
the portable switchboard. The units are operated by the switches 
attached directly to the equipments. This is done because there is 
always an operator at each unit. 

When remotely controlled boards are used an operator stands 
near the cameraman with the small control switch box at hand so as to 
be able to operate the lights as directed by the cameraman. In the 
case of the heavy-current arc units operated by an attendant, the 
gaffer signals by yelling "Hit No. 5," or "Kill No. 8." 




FIG. 7. Schematic diagram for portable switchboards. 

There are advantages and disadvantages to both systems. The 
first arrangement, with its remotely controlled boards permanently 
installed at the tops of the stages, greatly lessens the amount of 
cable, particularly upon the floor, where it is very much in the way. 
On the other hand, it does represent considerable investment, much 
of which is idle. 

A few electrical departments have adopted the practice of marking 
each cable end as well as connecting points at the switchboards 
red for positive, blue for negative, and white for neutral to facili- 
tate making correct connections, particularly because the arc equip- 
ment requires that the current flow be in the proper direction. This 
practice, however, is not general, although it would be well if it were. 



R. E. FARNHAM, Chairman 
W. C. KUNZMANN V. E. MILLER 

J. H. KURLANDER M. W. PALMER 

E. C. RICHARDSON 



C. STRUSS 
F. WALLER 



442 STUDIO LIGHTING COMMITTEE 

REFERENCES 

1 Report of the Studio Lighting Committee, /. Soc. Mot. Pict. Eng., XVII 
(Oct., 1931), No. 4, p. 645. 

1 Report of the Studio Lighting Committee, J. Soc. Mot. Pict. Eng., XVm 
(May, 1932), No. 5, p. 666. 

3 JOY, D. B., BOWDITCH, F. T., and DOWNES, A. C.: "A New White-Flame 
Carbon for Photographic Light," J. Soc. Mot. Pict. Eng., XXII (Jan., 1934), 
No. 1, p. 58. 

4 FARNHAM, R. E.: "Mazda Lamps for Color Photography," J. Soc. Mot. 
Pict. Eng., XXI (Aug., 1933), No. 2, p. 166. 

8 BUTTOLPH, L. J.: "High-Intensity Mercury and Sodium Arc Lamps," 
J. Soc. Mot. Pict. Eng., XXIV (Feb., 1935), No. 2, p. 110. 

6 MOLE, P.: "New Developments in Arc Lighting," /. Soc. Mot. Pict. Eng., 
XXH (Jan., 1934), No. 1, p. 51. 

7 KLIEGL, H.: "The New Klieglight," /. Soc. Mot. Pict. Eng., XXIII (Dec., 
1934), No. 6, p. 359. 

8 RICHARDSON, E. C.: "A New Wide-Range Spot Lamp." Presented at the 
Spring, 1935, Meeting of the Society, at Hollywood, Calif. ; to be published in a 
subsequent issue of the JOURNAL. 

9 MOLE, P. : "New Developments in Portable Gas-Electric Generators for 
Motion Picture Lighting," J. Soc. Mot. Pict. Eng., XXI (Nov., 1933), 
No. 5, p. 413. 



THE USE OF FILMS AND MOTION PICTURE EQUIPMENT 

IN SCHOOLS* 



MARION EVANS** 

Summary. The educational motion picture has come as a safeguard against the 
extreme verbalism in education of a few years ago, which often led to serious miscon- 
ceptions by the pupil of facts that can now be presented as true, unbiased reproductions 
of life itself. In judging the attributes of a good teaching film certain definite criteria 
should be followed, as regards content, emotional force, interest, manner of presenting 
the subject matter, etc. 

Edison challenged the educational world fifteen years ago with the 
prophecy that "soon the motion picture film would take the place of 
the text-book in the classroom." This statement created a great deal 
of controversial discussion on the part of educators, some of whom 
maintained that pictures never could take the place of the text-book 
but should always be regarded as supplementary or secondary to the 
book. This limited view was due to the fact that Edison's prophecy 
came at a time that was characterized as a brief period of extreme 
verbalism in education. Teachers then were worshipping the text 
as the Moslem did his Koran as the one and only means of enlighten- 
ment. Lessons consisted of reading, alternated by "chalk and talk" 
lessons by the teacher and parrot-like recitations in which the stu- 
dents repeated forced memorized facts that often had little or no 
meaning. Those were the days when the teacher taught geography 
by having the class repeat for a week in unison such phrases as, "The 
equator is an imaginary line running around the earth" and then was 
very much surprised when Johnnie wrote in his written examination, 
' 'The equator is a menagerie lion running around the earth. ' ' Johnnie 
did not have any idea what an imaginary line was ; but he did know 
what a menagerie lion was, so the lion became a part of the word pic- 
ture that he had mentally drawn. 

Fortunately for the youngsters, this period of verbal instruction was 
brief, for we know that pictorial records and visual images have been 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** San Diego City Schools. San Diego, Calif. 

443 



444 MARION EVANS [J. S. M. P. E. 

used to record knowledge and communicate ideas from the time of 
the stone age to our present machine age. We can trace evidences of 
the uplifting power of visual education in the crude animal picto- 
graphs upon cave walls, upon carved totem poles, in sculptured leg- 
ends and myths upon temple walls, in starward pointing obelisks, in 
rose and amber colored paintings upon cathedral windows, in murals 
in modern art galleries, and in the life-like motion pictures screened 
in schools and theaters today. After centuries of visualization in 
education has come the art of photography in all its forms the 
photograph, the stereograph, the glass slide, the still film, and the 
motion picture film. These visual aids now serve as a safeguard 
against verbalism, for although the book gives the interpretation of 
life as seen by the author and the painting represents the world as 
visualized by the artist, it is the photographic record that gives a 
true, unbiased reproduction of life itself. It alone allows the pupil 
to face it squarely, to study it, and to interpret its message and draw 
conclusions in terms of his own understanding. For this reason, the 
photographic record is a teaching tool directly fashioned to fulfill the 
needs of modern education, which aims to teach boys and girls how to 
study as well as what to study, and not only to memorize facts but 
how to weigh and handle facts. 

When we realize that the film now discloses a whole new world for 
observation and study, bringing the miracles of nature realistically 
to the student and revealing many of the long-hidden secrets of 
Mother Earth, we understand why the alert teacher is eager to in- 
clude the motion picture in her "kit of teaching tools." We under- 
stand also how Edison, who knew the possibilities of the new medium, 
had the vision to see that the film would take the place of the book in 
many lessons as the approach to learning. 

Looking into the modern classroom, we see how, through pictorial 
experiences, geography becomes the "great, wide, beautiful, wonder- 
ful world, " with its varied peoples and their activities. History is 
relived, so that pupils cross the barren plains with the early pioneers 
in covered wagons or follow a good-will messenger in an aerial flight 
encircling the globe today. In the science laboratory, microscopic 
photography makes visible tiny objects that the human eye can not 
see. Slow-motion films show movements and growth that the eye 
does not ordinarily observe, and, within a few minutes, depict the 
tedious processes of days, weeks, months, and even years such as 
the unfolding of a flower, or a river slowly digging a new channel in 



Nov., 1935] MOTION PICTURES IN SCHOOLS 445 

the earth. In the study of astronomy, the telescopic lens brings dis- 
tant objects, such as the stars, down to a focused point of observation 
upon the classroom screen, while the film sound record, by means of 
amplification, reproduces audibly sounds that the human ear can not 
hear. 

To fulfill the growing demand for motion picture films in education, 
school visual education departments are having to turn to many differ- 
ent sources. In the past, the chief sources for school film purchases 
have been educational divisions of commercial companies such as 
Eastman Teaching Films, Inc., and Bell & Howell Filmo Library; 
universities, including Harvard, Yale, and Chicago, which have pro- 
duced educational films; and firms like the General Electric Co., 
which have made available industrial subjects at low cost or free to 
schools. A few travelogue films made by the major motion picture 
studios have proved satisfactory, although some of the subjects are 
out-of-date by the time they are released to the non-theatrical field. 
Today, we are looking with hope and enthusiasm to a new field the 
great unexplored realm of the amateur and professional independent 
motion picture producer, from whom we expect to obtain a great 
wealth of artistic and accurately filmed photographic material. The 
shortening of the work-day and increase in leisure time are encourag- 
ing many creative and inventive workers to make photography a 
profitable hobby. 

The great improvement in amateur equipment during the past few 
years is making it possible to use devices that produce pictures of 
truly professional quality. This fact is proved by the high standard 
that is set in the various competitive amateur cinema contests, which 
are yielding beautiful and highly instructional films that school film 
libraries may now include in their purchases such films as Japanese 
Lullaby, Water, Korean Rice Farmer all prize-winners in an amateur 
cinema contest. 

In judging the attributes of a good teaching film such standards as 
the following act as guiding criteria: 

(1) Does the subject-matter appeal to native human interests? 

(2) Does it contain sufficient mental stimulus to be thought-provoking, prob- 
lem-raising, or problem-solving? 

(5) Does it have social values, and a positive emotional appeal which makes it 
elevating, healthful, and inspiring? 

(4) Are the titles brief and simple? 

(5) Is the continuity good, with the main points of the lesson clearly defined 



446 MARION EVANS [ J. S. M. P. E. 

in a unified and balanced presentation of the subject so that some specific learning 
may be effected? 

(6) Is it true, according to the nature of the theme being portrayed, whether 
realistic or fantastic? 

(7) Is the photographic quality clear and artistic? 

(8) Is it so edited as to conform to the span of attention and the comprehen- 
sion of one of the four grade levels, namely, kindergarten-primary, elementary, 
secondary, and adult? 

In addition to judging the quality of a film, another fact that must 
be taken into consideration is the type of motion picture to be se- 
lected. Should the film be silent, audible, or colored? We answer 
this question by saying that the purpose of the lesson should deter- 
mine the type of film best adapted to its use. 

For example, if we are striving for the emotional appeal and aes- 
thetic appreciation of color elements, or if, as in nature study and 
science, color is in itself an important phase of identification of the 
object, then we select color. On the other hand, if we wish merely to 
visualize action, growth of moving life or development of a process, 
then we can use black and white. As for sound, if musical interpre- 
tation, sound, or voice effects contribute to the main point of empha- 
sis of the lesson, the sound or talking film is chosen. As an example, 
it might be advisable to have a colored sound film on local birds, be- 
cause the calls of the birds are as much a means of identification as 
their coloring. However, a colored silent film on the subject of the 
butterfly would serve our purpose. 

Thus, it may be seen that, unlike the theatrical world, schools have 
refused to discard the silent film and substitute the talkie exclusively 
because we believe that each type of film has its specific contribution 
to make to education. We know there are great possibilities in 
bringing music of the highest caliber to all children, and in giving 
immortality to magnetic personalities of all ages. The talkie or lec- 
ture film may also bring master teachers to the most remote schools. 
Sound films are, therefore, of value in the study of music and folklore, 
drama, literature, and language, whereas the specific contributions 
of the silent film seem to be that it stimulates the imagination and 
intellectual faculties and is more favorable to creative contemplation 
on the part of the child. As it invites spontaneous comment and ques- 
tioning by the pupil, the silent film is to the teacher what the x-ray 
is to the physician an instrument that may be used, first, to diag- 
nose the needs and interests of individuals in the class, and then to 
solve their problems. The motion picture film may be used to intro- 



Nov., 1935] MOTION PICTURES IN SCHOOLS 447 

duce and arouse interest in a subject, as development material, or for 
reviewing a lesson. 

In order to insure the most desirable results from the use of class- 
room films, the motion picture equipment must be standardized and 
must be as carefully selected as are the films. Standards by which 
the apparatus is judged are safety, economy, durability, adaptibility, 
simplicity, portability, projection quality, and general efficiency. 
We have found that the 16-mm. projectors are the best suited for 
classroom use, judged by these items, and therefore recommend 
placing at least one portable motion picture machine as a part of the 
standard furnishings in each school building. One hundred per cent 
of the schools in San Diego are so equipped at this time. 

While the motion picture projectors are placed permanently in the 
schools, the films are circulated, upon the requests of teachers, from a 
central library or visual education department. A glimpse at the 
shelves of such a film library would show topics covering practically 
every subject of the curriculum. The distribution records would re- 
veal an ever-increasing demand with a turn-over of from 5000 to 
10,000 film showings per month for every 500 subjects in the film 
library. 

Motion picture appreciation is being developed in four ways in 
modern school systems: 

(1) The schools are providing carefully selected film experiences 
for all children from the kindergarten through the senior high school 
by furnishing through the visual education library educational films 
to illustrate daily classroom lessons. Such films are selected ac- 
cording to certain attributes which go to make a good teaching film. 

(2) Teachers are encouraging students to discuss good current 
pictures which they see in local theaters, such as David Copperfield, 
Midsummer Night's Dream, Great Expectations, Little Women. Not 
only are such discussions introduced as a regular part of English, 
drama, and public speaking courses, but such pictures as Our Daily 
Bread, As the Earth Turns, Gabriel over the White House, and The Presi- 
dent Vanishes, which portray important social and economic problems 
of today, are used as the basis of lessons in social science and current 
history courses. 

(3) Teachers who are organizing new materials for the course of 
study are now collaborating with the visual education department 
in collecting films, books, pamphlets, and outlines that may be used 
in motion picture appreciation courses to be introduced in the junior 



448 MARION EVANS 

and senior high schools. Such a course would follow the lines of the 
appreciation of art and appreciation of music classes. Screen classics 
representing the various classes of photoplays, such as comedy, 
musical, melodrama, travelogue, cartoons, etc., would be intimately 
studied, and definite standards for judging good films would be dis- 
cussed with the pupils. 

(4) The schools are endeavoring to provide for actual film pro- 
duction experience during school life by means of photography classes 
and photoplay clubs, especially in high schools. Opportunity is 
afforded these young people to gain experience in making motion 
picture films as well as still photographs, which are exhibited in their 
annual salons. 



SYMPOSIUM ON NEW MOTION PICTURE APPARATUS 



During the Spring Convention at Hollywood, Calif., May 20-24, 1935, a sympo- 
sium on new motion picture apparatus was held, in which various manufacturers of 
equipment described and demonstrated their new products aud developments. Some 
of this equipment is described in the following pages; the remainder will be published 
in subsequent issues of the Journal. 

A NEW HIGH-FIDELITY SOUND HEAD* 



F. J. LOOMIS AND E. W. REYNOLDS** 



The sound attachment, used in conjuction with a picture projector and its com- 
ponents, is a device employing the electrical and mechanical equipment necessary 
to scan a film sound record, translating it into electrical vibrations which are sub- 
sequently amplified and reproduced as a faithful counterpart of the original record 
sound. The struggle to develop a film record sound reproducer has extended over 
a number of years and this effort has produced a conventional type of machine 
which is in universal use. Along with other components, common to all, this con- 
ventional type makes use of a fixed sound gate, as shown in Fig. 1. 

Experience has demonstrated that fixed sound gates have required constant 
attention because there has always been the possibility of accumulation of wax 
and emulsion upon the polished surfaces of the film guide and pressure shoes, and 
it is evident that frictional resistance to the passage of film results in wear upon 
the film, gate, and sprocket. 

Another major problem, to which engineers have devoted much time and 
thought, is that of achieving a constant speed of the film at the sound scanning 
point. Attempts to attain constant speed have largely been concentrated upon 
causing the feed sprocket below the sound gate (Fig. 1) to revolve at a uniform 
velocity. Elaborate mechanical filters between the motor and the feed sprocket 
and various types of direct drive have been used, all of which were designed to 
produce uniform rotational velocity of the feed sprocket; but, with few exceptions, 
no effort has been made to eliminate the annoying ripple necessarily present in 
sound reproduction from film that is fed only by a sprocket past the sound scan- 
ning point. 1 

In spite of the study given to their many functions, sound reproducers have 
not kept pace with the advancement that has recently taken place in the recorder 
field, 2 and for that reason, we were called upon to develop and design a new sound 
attachment that could satisfactorily reproduce our high-fidelity recordings. 

*Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
**RCA Manufacturing Co., Camden, N. J. 

449 



450 



APPARATUS SYMPOSIUM 



[ J. S. M. p. E. 



This sound head consists of a housing containing a removable exciter lamp 
socket with its mounting ; an optical system for focusing the light upon the sound- 
track of the film; a device for controlling the film laterally; a free-running film- 
driven sound take-off drum to which is attached a rotary stabilizer, which causes 
the sound-track to pass the scanning beam at a constant speed; a photo-cell and 
its transformer for translating the light variations, caused by the sound-track of 
the film, into electrical pulsations ; and an electric driving motor with motor con- 
trol, together with other necessary components. 

Compact and unique in design, the main case, shown in Figs. 2 and 3, embodies 




FIG. 1. Showing the film path and the fixed sound gate. 



a great many features of construction that are bound to have a pronounced effect 
upon the extended life of the machine. Consisting of box sections, extending from 
either side of a center plate, the main case is cast of iron in one piece. The front 
section contains the exciter lamp and the film-handling compartments, and at the 
rear is a housing for the transformer assembly. All bearing mounting holes for 
the principal rotating shafts are machined in this single casting, assuring parallel- 
ism and permanent alignment. At one end of the case, provision is made for at- 
tachment of the sound head to a pedestal and on the other end is cast a circular 
projection which serves as the front end bell for the driving motor. Ample open- 



Nov., 1935] APPARATUS SYMPOSIUM 451 

ings for ventilating the exciter lamp compartment have been cast in the main case 
in such a manner as to preclude stray light. 
One of the leading features of this design is the use of precision ball bearings 




FIG. 2. Rear view of the main case, housing the 
transformer assembly and the drive gears. 

upon all rotating shafts, including the motor shaft. The bearing retainers and 
deflectors protect the bearings from dirt and insure the retention of lubricant. By 
properly mounting, protecting, and lubricating the modern precision ball bearings, 




FIG. 3. Front view of the main case, housing the 
exciter lamp and film-handling compartments. 

it is possible to gain a greater length of life and a much longer period of trouble- 
free operation than with any other type of bearing. Fig. 4 illustrates the compact- 
ness of the complete assembly. 
Because of the wide range of power supplies, it seemed desirable that the sound 



452 



APPARATUS SYMPOSIUM 



[J. S. M. P. E. 



head be designed for various combinations of voltage and frequency as well as for 
direct current. Of course, this interchangeability of drive required a special built-in 
arrangement allowing the use of certain standard parts upon all drives Each 
motor armature shaft has a worm, either cut integral with or fastened upon the 
shaft, and supported by two ball bearings; one slidable, mounted in the front bell 




FIG. 4. The complete assembly. 



of the motor and located upon the shaft as shown in Fig. 5; the other firmly 
mounted in the drive gear chamber of the sound head. The entire drive gear as- 
sembly, including the bearings, runs in a bath of oil, which is made possible by 
the use of an oil seal mounted behind the rear motor bearing. The rear motor 
bearing carries the thrust and the radial loads of the worm drive. The motor 
worm drives a worm-gear mounted upon the feed sprocket shaft which, through 



Nov., 1935] 



APPARATUS SYMPOSIUM 



453 



an external gear-train, drives the hold-back sprocket shaft, as shown in Fig. 6, 
with the projector located above the sound head. 

The exciter lamp socket (Fig. 7) carries one standard 10-volt, 5-ampere lamp, 




FIG. 5. Showing the method of mounting the motor. 

the base of which is gripped uniformly around its circumference by a spring chuck 
cut into the end of a threaded sleeve which has a sliding fit in the socket body. 
The upper knurled nut is tapered inside and, when tightened, compresses the chuck 




FIG. 6. External gear-train driving the projector, and the hold-back 

sprocket shaft. 

jaws uniformly upon the lamp base, eliminating the possibility of deforming the 
base and ruining the lamp. 

The filament is adjusted vertically by rotating the lower knurled nut, and the 



454 



APPARATUS SYMPOSIUM 



[J. S. M. P. E. 



sleeve is prevented from turning by a locking screw which is tightened after the 
correct vertical adjustment has been made. Two long pilot-pins mounted in the 
head carry the socket body and permit instant and exact replacement of a burned- 
out lamp by a spare prefocused exciter lamp assembly provided with each sound 
head. A handle is cast upon the socket body to facilitate withdrawing and re- 
placing the unit. This speed of replacement is especially important during a per- 
formance. The socket assembly, together with its mounting, is strongly made 
from die castings and machined brass parts, and is designed for long and satis- 
factory service. 

The essential parts of the optical system are a condenser lens, a mechanical slit, 
and an objective lens, all mounted in a hermetically sealed barrel. In designing 




FIG. 7. Showing exciter lamp assembly, optical system, 
and film -handling mechanism. 



the optical system, consideration has been given to keeping its interior free from 
oil and dirt. It is of the conventional type; the condenser lens collecting the light 
and illuminating the mechanical slit is placed as close as possible to the lamp; and 
the objective lens, in turn, throws a reduced image of the illuminated mechanical 
slit upon the film at the sound-track. The entire optical barrel is a sliding fit in 
the body casting which is accurately and firmly mounted in the sound head as 
shown in Fig. 7. 

A knurled nut fixed axially and operating upon a threaded section of the optical 
barrel, provides focal adjustment of the system. At one end, the body is slotted 
transversely and is supplied with a lock-screw, thus forming a clamping means for 
definitely holding the optical barrel in the focal position. Focusing and locking 



Nov., 1935] 



APPARATUS SYMPOSIUM 



455 



are easily done directly from the front. Because of manufacturing precision, no 
angular adjustment of the mechanical slit is necessary. 

After passing through the film, the modulated light-beam encounters a lens 
(Fig. 7) which is a combination of a condenser and a prism, and is condensed and 
deflected upward and outward to the target of a standard phototube. The photo- 
tube is mounted parallel to the drum and is shielded from extraneous light by a 
cover (Fig. 7) which serves also to hold the tube in correct position, both the cover 
and the phototube being instantly removable from the head. 




FIG. 8. Showing the film path through the sound head. 

To assure proper shielding, leads from the phototube socket pass through the 
center of a large cored hole in the main casting to the phototube transformer, 
which is resiliently wrapped and placed inside a thick steel shield. This complete 
unit in turn is surrounded by resilient material placed between it and the walls of 
the transformer compartment, which is cast in the main case. By the double 
resilient mounting and double shielding provided by the cast-iron walls of the 
main case and the steel container, all audible disturbances, mechanical and elec- 
trical, are eliminated from the transformer. 

A very accessible terminal board for connecting the amplifier cable is mounted 
upon the transformer support (Fig. 6), and a large cast-iron cover affords easy 



456 



APPARATUS SYMPOSIUM 



[J. S. M. P. E 



entrance into the compartment. Care has been taken to prevent oil from enter- 
ing the transformer compartment, the top and side openings having oil-tight 
covers; and because of this, deterioration of insulating material within the trans- 
former compartment is avoided. 
Special care has been taken so to arrange the drum, contact roller and sprockets 





FIG. 9. The rotary stabilizer, disassembled. 

that film, as it passes through, is bent into as wide curves as possible (see Fig. 8). 
The film strippers, although effective, do not interfere with threading, and can be 
solidly located in proper adjustment. A long bearing in the pad-roller support as- 
sures firmness of the pad-roller in all positions, and the new lock-plate permits 




FIG. 10. The stabilizer drum-shaft and sound gate. 



easy and positive adjustment and location of the pad-roller. These pad-rollers 
swing far enough away from the sprockets to make threading of the film through 
the head a very easy matter, and snap definitely into position. 

In order to attain constant speed of the film at the sound scanning point, it has 
been customary in the conventional sound head to make use of large rotating 



Nov., 1935] 



APPARATUS SYMPOSIUM 



457 



masses in connection with other apparatus, and to achieve the desired results by 
what might be termed the "brute force" method. 

However, all things are relative; and if the masses employed in the conven- 
tional sound heads were large, the disturbing forces were large also, so that the 
final result, in so far as constancy of speed was concerned, was seldom all that 
could be desired. 

In this new sound head the rotating masses employed are comparatively small 
but the disturbing forces are relatively insignificant, so that the final result is much 
superior to that arrived at in the conventional manner. By mounting the sound 




FIG. 11. The rear view of the machine, with the gear guard in place. 



drum shaft in free-running precision ball bearings, friction is reduced to such a 
minimum that it is possible for the film to drive the drum without appreciable 
tension. The driving strain is so slight that the film is never pulled taut, except 
at the start. The curves assumed by the film when leaving the sound drum and 
approaching the feed sprocket, when in operation, are clearly shown in Fig. 8. 
Flatwise compliance inherent in the curves of the film, between the sound drum 
and the feed sprocket, positively isolates the sprocket tooth frequency from the 
scanning point. It also takes care of the small irregularities of speed introduced 
into the feed sprocket by the drive gears. Were it not for this compliance, it 
would be quite impossibele to use the direct drive, with all its advantages, that 
has been built into this new sound head. Because of the lack of tension upon the 
film as it is fed through the sound head, the feed sprocket can be dimensioned to 



458 



APPARATUS SYMPOSIUM 



[J. S. M. p. E. 



accommodate maximum film shrinkage with no possibility of impairing the repro- 
duced sound. Also, there is assurance of prolonged feed sprocket life. 

It is apparent that the elastic film loop shown in Fig. 8 between the feed sprocket 




FIG. 12. The operating side of the machine, showing convenience 
of starting switch and framing knob. 



and the drum will absorb film speed irregularities introduced by the sprocket, 
and in order to utilize this valuable film loop it is necessary only to insure uniform 
rotation of the drum. The time-honored expedient for uniform rotation is a fixed 
flywheel. However, fixed flywheel control of the drum speed is unsatisfactory, 
because the flywheel continually hunts or oscillates with the springy film loop in 



Nov., 1935] APPARATUS SYMPOSIUM 459 

the same manner that a weight suspended from a coil spring will oscillate under 
the slightest disturbance. It might be suggested that sufficient friction drag be 
applied to the drum shaft to prevent or damp the oscillations, but when this is 
done the film is immediately stretched taut between the feed sprocket and the 
drum, and the valuable film loop is lost. 

It was therefore necessary to develop a rotational speed control for the drum 
that would not oscillate with the springy film loop nor pull the loop taut so as to 
destroy it. The device that was developed to fulfill the requirements is called, 
for want of a more appropriate term, a "rotary stabilizer." 

A number of years ago C. R. Hanna discovered that two rotating masses or fly- 
wheels coaxially mounted upon one spring-driven shaft would be critically damped 
or, in other words, would not oscillate with the driving spring if the assembly were 
constructed under certain conditions. These conditions were that the inertias of 
the two flywheels should be approximately in the ratio of 8 to 1 , that the small 
flywheel be rigidly fastened to the shaft, that the large flywheel be free-floating 
upon the shaft and driven only through a perfectly viscous connection, and that 
the spring elasticity and the viscous connection have a certain relation to each 
other and to the flywheels. The mathematical theory disclosing these conditions, 
developed originally by Hanna, is further elaborated and expanded by E. D. 
Cook. 3 

It is sufficient to note that the theory leads to a device for controlling the drum 
speed which exactly meets the two conditions, first, that it does not oscillate with 
the elastic film loop, and, second, that it does not pull the loop taut. 

The rotary stabilizer shown in Fig. 9 was designed, according to the theory, as 
follows: The light flywheel was constructed as a short cylindrical casing made of 
the lightest possible alloy and firmly fastened to the drum shaft, and the free-float- 
ing heavy flywheel was carried inside the casing upon a ball bearing mounted upon 
its hub. The viscous driving connection to the heavy flywheel is a light oil which 
completely fills the casing and surrounds the flywheel, and the spring drive to 
the assembly is the elastic film loop from the sprocket to the drum. The casing 
is, of course, hermetically sealed by a cover which retains the oil and excludes dirt 
from the assembly. 

The results attained with the device constructed as outlined were in accordance 
with the theoretical predictions, and the passage of a film splice or a severe manual 
disturbance of the film loop does not result in a single complete oscillation of the 
drum and rotary stabilizer. 

It is interesting to note that the theoretical proportions of the rotary stabilizer 
demand a construction that is at variance with earlier empirical designs of similar 
damping devices. In these devices the flywheel fastened to the shaft is very large 
in proportion to the free-floating damping flywheel, and the rotary stabilizer re- 
verses this ratio with greatly improved results. The lack of oscillation between 
the stabilizer and the film loop is due to the fact that the energy of the distur- 
bance passes from the film loop to the casing and is dissipated in the oil film between 
the casing and the flywheel. The proportional inertias of the casing and the fly- 
wheel are such that the small amount of energy stored in the light casing is in- 
sufficient to affect the rotation of the flywheel seriously. Over a period of years 
and in a large number of installations the rotary stabilizer has proved to be an 
extremely satisfactory, accurate, and trouble-free method of controlling film speed. 



460 APPARATUS SYMPOSIUM 

The drum and shaft (Fig. 10) are made from a one-piece chrome-nickel steel 
forging, heat-treated to assure sufficient rigidity. 

The film is controlled laterally at the scanning point by the specially designed 
guide and contact roller shown. 

In this assembly, the film passes between two flanged rollers mounted upon a 
common shaft. One roller is fixed in position, the other is slidable, being held 
against the rear edge of the film under slight spring pressure. Both flanged rollers 
are carried in a knee-jointed arm which swings upon a stud mounted in the main 
case, and lateral adjustment of the film is made by moving the entire assembly 
upon the stud. This assembly can be moved and locked in any desired position 
by rotating a split thumb-nut upon the front end of the stud, against which the 
assembly is firmly held by a spring located back of the lower arm. When the as- 
sembly is in the closed position, a resilient insert placed in the slidable half of the 
guide roller contacts the film firmly enough to assure proper traction upon the 
drum. This light contact is maintained at a fixed amount by a spring placed in 
the knee joint of the arm. Grooving the guiding flanges by the film is rendered 
impossible because of the rotation of the entire assembly upon the free-running 
ball bearings. 

In a project of this character some attention must be given to the appearance 
and safety of the mechanism. Fig. 11, showing the rear view of the machine, 
clearly depicts the gear guard which not only protects the operator from injury, 
but completely covers all moving parts and adds to the general symmetry of the 
design. A view of the operating side of the machine, Fig. 12, affords an idea of 
the accessibility of the motor-starting switch and the framing knob placed upon 
the end of the motor shaft for the convenience of the operator. 

Fully realizing the requirements of the motion picture theater and recognizing 
a sound-critical public, it has been our aim to create a mechanism which is sub- 
stantially better mechanically and capable of excellent performance. The machine 
will do justice to high-fidelity recordings, and give long, uninterrupted service, 
with minimum film wear. 

REFERENCES 

KELLOGG, E. W.: "A New Recorder for Variable-Area Recording," /. Soc. 
Mot. Pict. Eng., XV (Nov., 1930), No. 5, p. 653. 

'ZIMMERMAN, A. G. : "Film Recorders," /. Soc. Mot. Pict. Eng., XX (March, 
1933), No. 3, p. 211. 

3 CooK, E. D.: "The Technical Aspects of the High-Fidelity Reproducer," 
/. Soc. Mot. Pict. Eng., XXV (Oct., 1935), No. 4, p. 289. 



SOCIETY ANNOUNCEMENTS 



WASHINGTON CONVENTION 

As this issue of the JOURNAL goes to press at the time of the Fall Convention 
at Washington, D. C., details concerning the Convention will be published in the 
December issue. The Tentative Program of the Convention, which was mailed 
to all the members of the Society recently, will, as usual, be followed by the 
Final Program, which will be distributed at the Convention. For the benefit of 
those who were unable to attend the Convention, the Final Program and a de- 
scription of the highlights of the Meeting will be published in the next issue. 
The papers presented at the various sessions will be published in the JOURNAL 
during the next few months. 

STANDARDS COMMITTEE 

At a meeting held at the Hotel Pennsylvania, New York, N. Y., September 26th, 
the following items were considered by the Committee: (1) report on the Euro- 
pean 16-mm. sound film situation by Mr. G. Friedl, who attended the recent 
Berlin and Paris Congresses as representative of the Sectional Committee on 
Motion Pictures under the A. S. A., and from Dr. W. Clark, who attended the 
Paris Congress as chairman of the American National Committee of the Interna- 
tional Congress of Photography; (2) the relation of the S. M. P. E. Standards Com- 
mittee to the various other standardizing groups, foreign and domestic; (3) 
proposed revision of various drawings in the Standards Booklet, in order to clarify 
them and increase their usefulness; (4) possibility of standardizing camera, sound, 
and printer sprockets; (5) possible standardization of screen brightness; (6) 
consideration of the photoelectric cell standards proposed by the British Standards 
Institution; (7) proposed layout for 8-mm. sound film; (8) 16-mm. sound test 
film, similar to the present S. M. P. E. Standard 35-Mm. Sound Test Film; (9) 
dimensions and footage of reels, proposed by the Academy of Motion Picture Arts 
and Sciences; (10) possibility of supplying standard densities to studios and 
laboratories for checking their densitometric work; (11) proposed change of lead 
of the sound over the picture from the present standard of 25 frames to 26 frames, 
proposed at the recent Congress at Paris; (12) definition of safety stock. 

The various items listed above will be discussed in detail in the Report of the 
Standards Committee, to be presented at the Washington Convention and subse- 
quently published in the JOURNAL. 

BOARD OF GOVERNORS 

A special meeting of the Board of Governors was held at the Hotel Pennsyl- 
vania, New York, N. Y., September 13th, for the principal purpose of completing 
work upon "Administrative Practices," which had been begun nearly a year 
ago, but the completion of which was constantly interfered with in the meantime 

461 



462 SOCIETY ANNOUNCEMENTS [J. S. M. P. E. 

by current matters. "Administrative Practices" is a compendium of all current 
operating policies and procedures of the Society, instituted by the Board of 
Governors aside from the provisions of the Constitution and By-Laws, and is 
for the guidance and reference of the members of the Board in their work. It is 
to be continually kept up to date, as new policies or actions are taken by the 
Board at the various meetings. 



SOCIETY AWARDS 

Announcement of the recipients of the Journal Award and the Progress Medal 
Award will be made at the Semi-Annual Banquet of the Society at the Washington 
Convention, October 23rd. The Journal Award is given for the most outstanding 
paper originally published in the JOURNAL during the preceding calendar year; 
and the Progress Medal is awarded in recognition of any invention, research, or 
development which, in the opinion of the Progress Award Committee and the 
Board of Governors, shall have resulted in a significant advance in the develop- 
ment of motion picture technology. The Progress Medal Award, it will be 
noted, does not necessarily apply to work done only during the current year. 



PACIFIC COAST SECTION 

Election of officers of the Pacific Coast Section has just been completed, with 
the following results: 

G. F. RACKETT, Chairman 

H. W. MOYSE, Secretary-Treasurer 

C. W. HANDLE Y, Manager 

The fourth member of the Board of Managers is K. F. Morgan, Manager, 
whose term does not expire until December 31, 1936. Mr. E. Huse remains a 
member of the Board for another year, as Past- Chairman. 



2000-FT. REEL 

Proposals for the standardization of a 2000-ft. reel to take the place of the pres- 
ent 1000-ft. reel were recently made by the Academy of Motion Picture Arts and 
Sciences. The subject had previously been considered and reported on by the 
S. M. P. E. Committees on Projection Practice and Exchange Practice in the 
June, 1934, JOURNAL. 

Meetings of representatives of the various exchange companies were recently 
held for the purpose of considering these proposals and for gathering economic 
and technical data involved in changing from the present standard to the pro- 
posed one, in the offices of the Motion Picture Producers and Distributors of 
America, under the chairmanship of Mr. A. S. Dickinson. As a result of these de- 
liberations it appears that the industry is favorable toward making the change, if 
certain conditions are adhered to for the present, and a complete report on the 
subject will be presented by Mr. Dickinson at the Washington Convention, May 
21st. 



Nov., 1935] SOCIETY ANNOUNCEMENTS 463 

RESEARCH FELLOWSHIP 

As an extension of the researches carried on at the U. S. Bureau of Standards 
in the Paper Section, relative to the preservation of records on paper, a 
study was recently initiated on the stability of cellulose acetate motion picture 
films with respect to their use for the reproduction of record material. Dr. J. R. 
Hill, formerly associate chemist in the Examining Division, Civil Service Com- 
mission, and assistant chemist in the Bureau of Chemistry and Soils, was ap- 
pointed Research Associate by the sponsoring National Research Council to 
conduct the investigation. 

SOCIETY SUPPLIES 

Reprints of Standards of the SMPE and Recommended Practice may be obtained 
from the General Office of the Society at the price of twenty -five cents each. 

Copies of Aims and Accomplishments, an index of the Transactions horn Octo- 
ber, 1916, to June, 1930, containing summaries of all the articles, and author and 
classified indexes, may be obtained from the General Office at the price of one 
dollar each. Only a limited number of copies remains. 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 
JOURNAL, may be obtained from the General Office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



The Society regrets to announce the death of 

WILLIAM KENNEDY LAURIE DICKSON 

Honorary Member of the Society 

September 30, 1935 



STANDARD S. M. P. E. 

VISUAL AND SOUND TEST REELS 

Prepared under the Supervision 

OF THE 
PROJECTION PRACTICE COMMITTEE 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 
Reels sold complete only (no short sections). 

PRICE $37.50 FOR EACH SECTION, 
INCLUDING INSTRUCTIONS 

(Shipped to any point in the United States) 

Address the 

SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL PENNSYLVANIA 
NEW YORK, N. Y. 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXV DECEMBER, 1935 Number 6 



CONTENTS 

Page 

Proceedings of the Semi- Annual Banquet at Washington, D. C., 

October 23, 1935 467 

Citation of Thomas Armat G. E. MATTHEWS 468 

The Work of Drs. L. A. Jones and J. H. Webb 

E. A. WILLIFORD 473 

The Work of Edward Christopher Wente. J. I. CRABTREE 478 

The New Era in Motion Pictures W. H. HAYS 483 

Analysis of the Distortion Resulting from Sprocket-Hole Modu- 
lation E. W. KELLOGG AND H. BELAR 492 

The Calibrated Multi-Frequency Test-Film F. C. GILBERT 503 

Uniformity in Photographic Development J. CRABTREE 512 

A Consideration of Some Special Methods for Re-Recording .... 
E. D. COOK 523 

Report of the Committee on Non-Theatrical Equipment 541 

Highlights of the Washington Convention 545 

Program of the Washington Convention 549 

Society Announcements 553 

Author Index, Vol. XXV, July-December, 1935 557 

Classified Index, Vol. XXV, July-December, 1935 560 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREB, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President : WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York. N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF. 250 W. 57th St.. New York, N. Y. 



PROCEEDINGS OF THE SEMI-ANNUAL BANQUET OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



WARDMAN PARK HOTEL, WASHINGTON, D. C. 
OCTOBER 23, 1935 



About two hundred members and guests of the Society assembled 
at the Fall, 1935, Semi- Annual Banquet at the Wardman Park Hotel, 
Washington, D. C., on October 23rd. Guests at the speakers' 
table included: Mr. Will H. Hays, President of Motion Picture Pro- 
ducers and Distributors of America, Inc., and speaker of the evening; 
The Honorable Kent Keller, Chairman of the Library Committee of 
Congress; The Honorable Justyn Miller, Special Assistant Attorney 
General of the United States; Dr. Lyman J. Briggs, Director of 
the U. S. Bureau of Standards, Washington, D. C.; Dr. R. D. W. 
Connor, First Archivist of the United States; Dr. John G. Frayne, 
Electrical Research Products, Inc., Hollywood, Calif.; and Mr. 
J. I. Crabtree, Editorial Vice-President of the Society of Motion Pic- 
ture Engineers. 

After an hour of dinner-dance music the guests at the speakers' 
table were introduced by President Tasker, who addressed the 
gathering as follows : 

"The Society of Motion Picture Engineers, organized in 1916 for 
advancement in the theory and practice of motion picture engineering 
and the allied arts and sciences, has devoted itself to the cause of 
'Progress' from that day to this. As time passed, the Board of 
Governors felt it expedient to establish certain awards which would 
tend to stimulate progress, and to that end there have been created 
the grade of Honorary Membership, the Honor Roll, the Journal 
Award, and the Progress Medal Award. 

"Pioneers of the motion picture industry who have contributed sub- 
stantially to the progress of the art have, from time to time, been 
awarded recognition by the Society in the form of Honorary Member- 
ship. Upon the demise of these notable pioneers, the Society has 
made provision for placing their names upon the 'Honor Roll,' 
which appears regularly in the JOURNAL of the Society every month. 
Honorary Members who now survive are : 

467 



468 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. p. E. 

FREDERIC E. IVES 
Louis LUMIERE. 

"The Presidencies of the Royal Photographic Society, The Society 
Franchise de Photographic, and the Deutsche Kinotechnische Gesell- 
schaft have also been accorded Honorary Memberships. Those 
names which now appear upon the Honor Roll of the Society, all of 
whom were Honorary Members, with the exception of the first two, 
since the Honor Roll was not established until after their deaths, 
are: 

Louis AIME AUGUSTIN LE PRINCE 

WILLIAM FRIESE-GREENE 

THOMAS ALVA EDISON 

GEORGE EASTMAN 

JEAN ACME LE ROY 

C. FRANCIS JENKINS 

EUGENE AUGUSTIN LAUSTE 

WILLIAM KENNEDY LAURIE DICKSON 

"At the Hollywood convention of the Society in May, 1935, Honor- 
ary Membership was conferred upon Mr. Thomas Armat of Wash- 
ington, D. C., for his pioneer contributions to the motion picture art. 
Presentation of the scroll of Honorary Membership was deferred in 
order that it might be personally presented to Mr. Armat this even- 
ing, on the occasion of our semi-annual banquet. In connection 
with any proposal for Honorary Membership, it is the duty of the 
Historical Committee of the Society to study the historical records 
and documentary evidence, and to prepare a statement of the 
grounds upon which the candidate may be properly considered for 
Honorary Membership. It is therefore my pleasure to call upon 
Mr. Glenn E. Matthews, a member of the Historical Committee, for 
appropriate citation of Mr. Armat's contributions to the motion pic- 
ture art." 

CITATION OF THOMAS ARMAT 

GLENN E. MATTHEWS 

Forty-two years ago, while attending the World's Fair at Chicago, 
a young engineer named Thomas Armat visited the exhibit of the 
Anschutz tachy scope. This device consisted of a small disk around 
the rim of which was mounted a series of glass transparencies made 
from a group of photographs taken in rapid succession. Behind the 
wheel at the top was a Geissler tube, which furnished the light to 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 



469 



illuminate one picture. The device was placed inside a box which 
had a peep-hole opening located opposite the illuminated trans- 
parency, where the scene was viewed by one person at a time. As 
the wheel rotated the light flashed, and an illusion of motion was 
produced. Describing this experience, Mr. Armat said, "The idea of 




Thomas Armat. 

bringing scenes from far distant and interesting countries and pro- 
jecting them upon a screen before comfortably seated spectators, was 
an exciting thought." 

A year later in Washington, D. C., Mr. Armat saw the Edison 
kinetoscope, and shortly afterward, during the summer, he began 
experimental research on projector design. To improve his knowl- 
edge of electric arc illumination, he enrolled in the Bliss School of 



470 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. p. E. 

Electricity, in the Fall of 1894, where he made the acquaintance of 
C. Francis Jenkins, founder of our Society, who, he learned, was also 
interested in the subject of projection of pictures. 

Armat and Jenkins joined forces in March, 1895, and constructed 
as their first model a projector, after the peep-hole kinetoscope con- 
tinuous motion principle but employing a different method of il- 
luminating the gate. The machine was unsuccessful, and under 
Mr. Ararat's supervision a second device was produced which is said 
to have been "the first projecting machine ever made that embodied 
an intermittent movement with a long period of rest and illumination 
of the pictures on the film." 

Joint patent protection was requested, and later granted (July, 
1897) on this apparatus, which, however, turned out to be a mechani- 
cal failure. The heavy sprocket and mutilated gear that was used 
were soon battered out of shape. Although satisfactory in some re- 
spects, Armat regarded the machine as uncommercial, and addressed 
himself to the task of devising a practicable one. 

The third machine was built in August, 1895, and utilized a modifi- 
cation of the Demeny beater type of intermittent movement. The 
temporary model, conceived by Armat, proved satisfactory, and a 
more substantial model was constructed immediately with which 
several successful exhibitions were given in his office in Washington. 
The projector was used also for several public showings at Atlanta, 
Georgia, at the Cotton State Exposition in September, 1895. Two 
duplicate machines were made and shipped to Atlanta. Early in 
October, 1895, Armat made further improvements in his machine, 
including the introduction of one of the most fundamental elements of 
a successful intermittent projector mechanism the "loop" or slack- 
forming device. 

Films for all models of the projector were those made by Thomas 
A. Edison and supplied through his agent, Raff & Gammon of New 
York. The success of the projector led Mr. Armat to contact Raff & 
Gammon, to find out whether they would be interested in the appara- 
tus. A demonstration satisfied them, and a showing before Mr. 
Edison was arranged at the Edison plant at Orange, N. J., in Febru- 
ary, 1896. As a result, a contract was signed between the parties 
concerned to manufacture eighty machines to be leased out on a 
royalty basis. 

The first public demonstration with one of these machines took 
place before a large audience in Koster and Bial's Music Hall in 



Dec., 1935] PROCEEDINGS OF SEMI- ANNUAL BANQUET 471 

New York on the evening of April 23, 1896. Each scene was ap 
plauded enthusiastically, especially one showing the storm-tossed 
waves breaking over the pier at Dover, England, a scene that was 
made by another pioneer of the motion picture, Robert Paul. 

Thus was launched one of the precursors of the modern motion 
picture projector. It was named by Mr. Armat the "Vitascope," 
and marketed by agreement as the Edison Vitascope. In September, 
1896, a patent was filed on the use of the Geneva cross movement, 
which greatly improved the intermittent action of the projector. 
This patent was issued to Mr. Armat in March, 1897. 

These are the principal details in the story of the development of 
the Vitascope by one of the noteworthy pioneers of this great industry. 
Mr. President, I have the honor to present the name of Thomas 
Armat for formal recognition by this Society. 



Amid enthusiastic applause from the audience, Mr. Armat ap- 
proached the speakers' table, where President Tasker presented to 
him the scroll of Honorary Membership, addressing him as follows: 

"Mr. Armat, no mere words of mine could increase the pleasure 
or importance of this great occasion, nor add to our high esteem of the 
contributions that you have made to the early development of this 
art which is so near to the heart of all of us. I deem it an honor and a 
privilege to award to you this certificate of Honorary Membership." 

Mr. Armat responded, "I wish to thank the Society of Motion 
Picture Engineers for bestowing upon me this honor. I shall always 
cherish this scroll of Honorary Membership and will endeavor in 
every way possible to support the ideals and further the purposes of 
the Society of Motion Picture Engineers." 

When Mr. Armat had resumed his seat, amid further applause, 
President Tasker continued as follows : 

"Realizing that Honorary Membership is necessarily limited to the 
recognition of the pioneers of the industry, the Society gave thought 
to means that might stimulate and reward present-day workers in this 
field. In consequence, there was recently established the "Journal 
Award," consisting of a prize of fifty dollars accompanied by an 
illuminated certificate of award, to be presented annually to the 
author or authors of the most outstanding paper originally published 
in the JOURNAL of the Society during the preceding calendar year. 
The first such award (1933) was granted to Dr. Peter Andrew Snell of 



472 



PROCEEDINGS OF SEMI-ANNUAL BANQUET (J. S. M. P. E. 



Rochester, New York, for his paper entitled "An Introduction to 
the Experimental Study of Visual Fatigue," which was published in 
the May, 1933, issue of the JOURNAL. 

"The research upon which Dr. Snell based his paper was done at the 
University of Rochester under the S. M. P. E. Fellowship established 
by Mr. George Eastman, then Honorary Member of the Society. 



:IETYOF MOTION RCTURE ENGINEERS 



FOUNDED 



<'|HM|' 

lB| 
THIS IS TO CERTI FY THAT f H E PAPER ENTITLE D 



! NCOftPOR ATED 

1916 



HAS BEEN DESIGNATED - BY THE -JOURNAL -AWARDS 
COMMITTEE AS THE -MOST OUTSTANDING PAPER 
ORIGINALLY PUBLISHED IN THE JOURNAL OF 
THE -SOCIETY -OF -MOTION PICTURE -ENGINEERS 
DURING THE YEAR - ONE THOUSAND NINE 
HUNDRED -THIRTY -.THREE 



ARDS COMMITTff. 



Journal Award Certificate. 

Unfortunately, Dr. Snell passed away soon after the publication of 
his paper, and so it happened that a year ago the fifty dollar prize was 
presented posthumously to his widow. At that time the Journal 
Award was so recently established that the design of the illuminated 
certificate was not yet finished. It was to have been our pleasure to 
present this certificate this evening to the parents of Dr. Peter Snell, 
but this afternoon I received from them the following telegram: 
'Kindly express to members of the S. M. P. E. our sincere apprecia- 
tion of the honor bestowed upon Dr. Peter A. Snell. We regret our 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 473 

inability to attend the banquet and will greatly appreciate it if 
Mr. Crabtree will receive the certificate for us. Sincerely, Dr. & Mrs. 
Albert C. Snell.' 

"Mr. Crabtree, I feel that it is most appropriate that you who are 
responsible for the publication of the JOURNAL of the Society of 
Motion Picture Engineers, and for the establishment of the S. M. P. E. 
Fellowship under which Dr. Snell worked, and further, through whose 
diligent efforts this beautiful and symbolic Journal Award certificate 
was created, should now be the one to receive this first certificate of 
its kind on behalf of Dr. and Mrs. Albert C. Snell." 

Mr. Crabtree replied as follows: "I am sure that Dr. and Mrs. 
Snell will greatly appreciate this beautiful certificate which has been 
awarded to their son, Dr. Peter Snell, and I shall be very happy to 
convey it to them." 

President Tasker: "A year has passed since the original award to 
Dr. Snell and it is now my very great pleasure to announce that the 
Journal Award Committee, after careful study, has recommended 
that the 1934 Journal Award be granted to Dr. Loyd Ancile Jones 
and Dr. Julian Hale Webb, for their paper entitled "Reciprocity Law 
Failure in Photographic Exposure." It is my pleasure to call upon 
Mr. E. A. Williford, a member of the Journal Award Committee, for 
appropriate citation." 

THE WORK OF DRS. L. A. JONES AND J. H. WEBB 

E. A. WILLIFORD 

The award for the most outstanding paper published in the JOURNAL 
of the Society for the year 1934 has been made to Drs. Loyd Ancile 
Jones and Julian Hale Webb for their paper entitled "Reciprocity 
Law Failure in Photographic Exposure," published in the September, 
1934, issue. 

Dr. Jones and his co-workers have investigated the subject of 
reciprocity failure for over a decade and have published no less 
than eleven papers of which the award paper was one. It is well 
known that photographic exposures vary over a wide range, depend- 
ing upon the nature of the subject being photographed and the inten- 
sity of the light. For example, a picture of a star through a telescope 
may require several hours' exposure, whereas the average snapshot 
in daylight requires only one-fiftieth of a second. According to the 



474 



PROCEEDINGS OF SEMI-ANNUAL BANQUET [J. S. M. p. E. 



reciprocity law, an exposure for which the product of the intensity 
of the light and the time of exposure is the same should produce the 
same photographic results, other factors being constant. The papers 
by Dr. Jones and his co-workers have shown, in general, that photo- 
graphic materials do not obey such a law. In the paper for which 




Loyd Ancile Jones. 



the award was made, a useful method of interpreting reciprocity 
data was described, which makes it possible to apply such data effec- 
tively for use with motion picture film. It was also shown that 
motion picture films are generally being used to utilize their char- 
acteristics to the maximum advantage. 

Dr. Jones has been chief physicist of the Kodak Research Labora- 
tories since 1916. He received an Electrical Engineering degree 



Dec., 1935] PROCEEDINGS OF SEMI- ANNUAL BANQUET 475 

from the University of Nebraska in 1908 and a Master of Arts degree 
in 1910. The University of Rochester honored him with a Doctorate 
in Science in 1933. Prior to entering the Kodak Laboratories in 
1912 he was engaged as a physicist at the U. S. Bureau of Standards 
from 1910 to 1912. During the World War in 1917-18, he was 




Julian Hale Webb. 

commissioned a Lieutenant in the U. S. Naval Reserve Force in 
charge of camouflage investigation. 

Dr. Jones has been active for many years in the Society of Motion 
Picture Engineers, having served as its President from 1923-26 
and its Engineering Vice-President since 1933. The Optical Society 
of America honored him by naming him as their President in 1930-31, 
and he is an active member of a number of other scientific societies. 



476 PROCEEDINGS OF SEMI-ANNUAL BANQUET [J. S. M. P. E. 

He has conducted and published extensive investigations in the fields 
of photometry, physical optics, illumination, colorimetry, physics of 
photography, visual sensitometry, and motion picture engineering. 

Dr. Julian Hale Webb received his Bachelor of Science degree in 
Electrical Engineering from Clemson College, South Carolina, in 1923, 
and his degrees of Master of Science in 1925, and Doctor of Philosophy 
in 1929, from the University of Wisconsin. From 1925 to 1929 Dr. 
Webb was assistant in the Physics Department at Wisconsin Uni- 
versity, and instructor of physics at Williams College from 1920 to 1931. 
He joined the Kodak Research Laboratories in 1931 and has special- 
ized in theoretical electrostatics and photographic theory. He is a 
member of the American Physical Society and the Optical Society of 
America. Mr. President, I have the honor to present Dr. Loyd 
Ancile Jones, who will receive the joint award in the absence of 
Dr. Julian Hale Webb." 



Amid prolonged applause Dr. Jones approached the speakers' 
table and received the Journal Award certificates for himself and 
for Dr. Webb, and responded as follows : 

"Members of the Society and friends, it is said that the genius of a 
great executive lies in the selection of his associates. Perhaps it 
could also be said that the genius of a research worker consists in the 
fortunate selection of very able co-workers. If this be the case, then, 
with Dr. Julian Hale Webb as the example, I am sure that I am a 
genius!" 

Mr. Tasker: "It is a requirement of the Journal Award that 
honorable mention be made of five other outstanding papers originally 
published in the JOURNAL during the corresponding year. The follow- 
ing are the papers thus given honorable mention : 

"On the Realistic Reproduction of Sound with Particular Reference to Sound 
Motion Pictures," H. F. Olson and F. Massa. 

"Sound-Film Printing," J. Crabtree, Bell Telephone Laboratories, New York, 
N. Y. 

"Stroboscopic-Light High-Speed Motion Pictures," H. E. Edgerton and K. J. 
Germeshausen, Massachusetts Institute of Technology, Cambridge, Mass. 

"Further Investigations of Ground-Noise in Photographic Sound Records," 
O. Sandvik, V. C. Hall, and W. K. Grimwood, Eastman Kodak Co., Rochester, 
N. Y. 

"Direct-Current High-Intensity Arcs with Non-Rotating Positive Carbons," 
D. B. Joy and A. C. Downes, National Carbon Co., Cleveland, Ohio. 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 



477 



"Now, it is a simple matter to decide that a Progress Medal shall 
be awarded, but to create a beautiful and appropriate medal is far 
from being simple. This task the Board of Governors imposed upon 
Mr. J. I. Crabtree, through whose efforts a number of very beautiful 
preliminary designs were submitted. One design, the work of 
Mr. Alexander Murray of the Eastman Kodak Company, Rochester, 
was particularly outstanding in its beauty and symbolic significance. 
Mr. Murray was asked to complete the design, which he has since 
generously donated to the Society; whereupon dies were made, the 
medal struck, and I hold in my hand the beautiful result. Since I 





The Progress Medal awarded to Dr. Edward Christopher Wente. 

can not pass it around for each of you to see, a photograph of the 
medal will be thrown upon the screen. 

"Referring first to the reverse face of the medal, the central hori- 
zontal panels afford opportunity to designate the name of the medalist 
and the purpose of the award. They also carry a number of little 
triangular elevations, which many of you will recognize at once as 
bromide crystals. Above the inscription appears an H&D curve, 
symbolic of the classical researches of Hurter and Driffield, to whom 
the industry is indebted for clarifying the photographic basis of 
successful motion picture photography, both of sound and of scene. 
In curved panels to the left and right appear sine waves, symbolic 
both of sound and light, which it is our modern purpose to imprison 
and again release for the enjoyment of a world-wide audience. An 
outermost circular panel bears the name of the Society. 

"Turning now to the obverse, we find that the center is a 



478 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. P. E. 

replica of the official emblem of the Society, itself inspired by the 
motion picture reel. Above and around this emblem are embossed 
the words 'For Progress,' and below are laurel branches, symbolic of 
achievement. 

"Surrounding the central portion of the design, a circle of film per- 
forations form a decorative motif, which cooperates symbolically 
with what, to my mind, is the most unique and significant feature that 
I have observed in a medal of this sort. It is, in fact, a reproduction 
of the earliest known bit of motion picture photography, the work of 
the early French scientist, Eugene Marey. 

"Much care was given by the Progress Award Committee, under 
the Chairmanship of Dr. Alfred N. Goldsmith, to the task of selecting 
the recipient of this, the first such award of the Society. At first it 
appeared that the task would be one of exceedingly great difficulty 
because of the enormous progress and the many contributions that 
had been made in recent years. Yet, before long, it became very 
clear indeed that one man stood out above the others in the im- 
portance and volume of his contributions to the motion picture art. 
This prolific worker was Dr. Edward Christopher Wente of the Bell 
Telephone Laboratories. His selection by the Progress Award 
Committee has been confirmed by the Board of Governors and it is to 
him, therefore, that the first Award of the Progress Medal will be 
made. 

"It is particularly fitting that the citation of Dr. Wente's contribu- 
tions to the motion picture art should be made by the man whose 
untiring efforts brought about the creation of this beautiful and 
impressive Progress Medal. It therefore gives me a great deal of 
pleasure to call upon our Editorial Vice -President, Mr. J. I. Crabtree, 
for appropriate citation of Dr. Wente's researches and inventions in 
the motion picture field." 



THE WORK OF EDWARD CHRISTOPHER WENTE 
J. I. CRABTREE 

For the past twenty years Dr. Edward Christopher Wente has 
been engaged at the Bell Telephone Laboratories and its predecessor 
organization in acoustical problems and in the development of special 
types of acoustical devices. One of the early problems upon which 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 479 

he worked was the development of a high-quality microphone that 
would translate sound into corresponding electrical currents with a 
degree of fidelity not previously approached. Microphones available 
at the time, while successful in telephony, were unsatisfactory for the 
transmission of music or the more subtle characteristics of the voice, 




Edward Christopher Wente. 

upon which the success of actors and public speakers so largely de- 
pended. This problem was successfully met in the condenser micro- 
phone, which depends for its action upon the change of electrical 
capacity between a metal diaphragm and a metal plate as the dia- 
phragm vibrates. Microphones of this type are used extensively 
at the present time not only for recording sound for motion pic- 
tures but also for broadcasting and recording phonograph records. 



480 PROCEEDINGS OF SEMI-ANNUAL BANQUET [J. S. M. P. E. 

They have been invaluable in fundamental studies of the characteris- 
tics of speech and music, and are part of the international refer- 
ence system for telephony. The John Price Wetherill Medal was 
awarded to Dr. Wente by the Franklin Institute in recognition of this 
work. 

In 1931 he completed the development of another high-quality 
microphone of entirely different type, known as the moving-coil 
microphone. In this device the voice currents are generated in a 
delicate coil of wire which is attached to the diaphragm and vi- 
brates in a magnetic field. This microphone has also found extensive 
application recently in recording sound for motion pictures and in 
broadcasting. 

The production of high-quality microphones required the develop- 
ment of a number of auxiliary devices, in particular, means for mea- 
suring their performance quantitatively. For this purpose Dr. Wente 
devised an alternating-current potentiometer and a thermophone to 
apply a known sound pressure to a microphone diaphragm. No satis- 
factory means for carrying out this necessary procedure in measuring 
the operating characteristics of microphones was previously available. 

The study of recording sound upon photographic film by the vari- 
able-density method, which is the type of record now used by many 
of the major motion picture producing companies in this country, has 
occupied a large part of Dr. Wente's time since 1920. This led to 
the development of the light- valve, which provides means for varying 
the light falling upon the film in accordance with the current gener- 
ated by the microphone. The device is now exclusively used by all 
Western Electric Company licensees in the production of sound 
pictures. 

Another essential element in the reproduction of sound is the loud 
speaker. In this field Dr. Wente introduced new principles of de- 
sign which resulted in a loud speaker more than fifty times as efficient 
and with at least one hundred times the sound output capacity of 
those previously available. Still more important, it gives a much 
more faithful sound reproduction. This loud speaker was used in 
the first showing of sound pictures by the Western Electric System, 
through the Vitaphone Corporation of the Warner Brothers in 1926, 
and still constitutes the sound reproducer in a large proportion of 
the theaters of the country. 

Control of acoustics both in the studio and the theater is of great 
importance in determining the quality of the sound that reaches the 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 481 

audience. Dr. Wente has made valuable contributions in studying 
these problems also. One of the factors that affect the acoustical 
characteristics of a room is the reverberation time, or the rate at 
which sound builds up and decays. To measure this he devised an 
electrical method which has completely supplanted the older and less 
accurate methods both here and abroad. More recently he has 
developed another device known as the high-speed sound level re- 
corder, which operates with such extreme rapidity that it has made 
available entirely new information upon the acoustical characteristics 
of rooms. 

The transmission of music by the Philadelphia Orchestra in audi- 
tory perspective from Philadelphia to Washington in 1933 was carried 
out with microphones and special loud speakers developed by Dr. 
Wente. This demonstration showed not only that the music of a 
large orchestra could be carried over telephone circuits many miles 
and reproduced so faithfully as to be indistinguishable from the 
original, but that the range of loudness under the control of the 
director at the receiving end could be increased to many times that of 
the orchestra itself, thereby permitting previously unattainable emo- 
tional effects. 

The results of his investigation in sound recording, communica- 
tion engineering, and acoustics have been published in various scienti- 
fic and technical journals, including those of the Society of Motion 
Picture Engineers, the American Institute of Electrical Engineers, 
the Acoustical Society of America, and also in the Physical Review, 
American Architect, and the Bell System Technical Journal. 

Dr. Wente was graduated from the University of Michigan in 
1911, after which he received a degree in Electrical Engineering at the 
Massachusetts Institute of Technology in 1914 and that of Ph.D. 
from Yale University in 1918. He was elected a member of the 
honorary society of Sigma Xi during his senior year at college, and 
held the John Sloane Fellowship in physics at Yale University. He 
taught physics at the University of Michigan while a student and 
was instructor in physics and mathematics at Lake Forest College 
from 1911 to 1912. In 1914 he began his association with the Re- 
search Division of the Engineering Department of the Western Elec- 
tric Company, the predecessor of the Bell Telephone Laboratories. 
Dr. Wente is a Fellow of the American Physical Society, the Ameri- 
can Association for the Advancement of Science, the Society of 
Motion Picture Engineers, and the Acoustical Society of America. 



482 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. P. E. 

He has been a member of the Editorial Board of the Acoustical 
Society since its organization, and is at present a member of its Execu- 
tive Council. 

With the exception of two years spent in graduate study at Yale 
University, Dr. Wente has been continuously engaged since 1914 
in technical research work in the Bell System. He has displayed 
rare inventive ability and has had exceptional success in encouraging 
the efforts of others by fruitful suggestions in a difficult field of work. 
This work has been related for the most part to acoustics and acousti- 
cal instruments, with special reference to their application to the 
recording, transmission, and reproduction of speech and music. 



As Mr. Crabtree concluded, the audience with one accord arose to 
its feet, and amid loud applause Dr. Wente approached the 
speakers' table, where the Progress Medal was presented by Presi- 
dent Tasker with the words: "Dr. Wente, I account it a very great 
honor, which I have not earned, to bestow upon you this honor which 
you so gloriously have earned." In response Dr. Wente said: 

"Thank you, Mr. President. I want to express to the Society my 
great appreciation of this honor, coming from an organization that 
can justly be proud of the many outstanding technical achievements of 
its members. I assure you that this beautifully embellished medal 
will be a most cherished possession." 

President Tasker: "As you are all aware, the feature of this 
evening's program is to be an address by Mr. Will H. Hays, Presi- 
dent of the Motion Picture Producers and Distributors of America, 
Inc., which will be broadcast over the facilities of the National 
Broadcasting Company. The hour of the broadcast is at hand, and 
there remains just time for us to express our deep appreciation to 
the many firms and individuals who have made possible the tre- 
mendous success of this thirty-eighth convention of the Society of 
Motion Picture Engineers. 

"March 11, 1922, is a date with a permanent identity and a lasting 
significance in the motion picture industry. On that date was or- 
ganized the Motion Picture Producers and Distributors of America, 
Inc., with Mr. Will H. Hays as President of the Association. So 
strongly has his personality become imprinted upon its activities that 
its original name is rarely heard. It is known throughout the world 
as 'The Hays Organization.' 



Dec., 1935] PROCEEDINGS OF SEMI-ANNUAL BANQUET 483 

"In his fourteen years of stewardship of this industry, innumerable 
issues and questions have arisen. It has required sound leadership, 
tireless activity, tact, and skill to deal with the diverse interests of 
many men and many minds, both inside and outside the motion pic- 
ture industry. I know of no man who could have brought to this 
position a greater aptitude in solving our complex and divergent 
problems. His ability to lay down those sound principles by which 
the motion picture industry governs itself, his success in meeting 
any new difficulty, the wideness of his vision, and the courage that 
has ever kept him upon the straight road to his ideals have placed 
him at the forefront of all motion picture improvements. 

"No art has ever depended so much upon science as the art of 
motion pictures. Steady progress in scientific and technical fields 
has made the motion picture undergo many readjustments. The 
advent of sound, the use of color, the betterment of camera and pro- 
jection technic have come swiftly upon us, and leadership and genius 
in meeting these advances were essential. In all the technical de- 
velopments of the industry, Mr. Hays has given of his time and 
ability, and has recognized that the engineering progress of the indus- 
try must go apace with its artistic and cultural advancement. 

"The achievements of Will Hays have been many and far-reaching. 
He has gained national recognition for his genius in organization and 
leadership, and he has won the confidence and esteem of all who have 
come in contact with him. 

"Ours is more than a business, far more than an industry; above 
everything else, it is a servant of happiness, of enlightenment, of 
culture, of human understanding. With the greatest of pleasure, 
I present to you tonight, a man who has served his country well and 
who for fourteen years has been the distinguished leader of our in- 
dustry: the President of the Motion Picture Producers and Dis- 
tributors of America, Will H. Hays." 

THE NEW ERA IN MOTION PICTURES 
WILL H. HAYS 

I am a movie fan. Fourteen years of most intense application to 
the problems and realities of this complex art-industry have increased 
this enthusiasm. I like them as the most ardent youngster likes 
them. I like them for the happiness they bring; for the relief they 
afford; for the information and knowledge and inspiration they carry; 
for the sheer service they accomplish as they lift a tired human being 



484 PROCEEDINGS OF SEMI-ANNUAL BANQUET [J. S. M. P. E. 

out of his fatigue and rebuild him with the magic of entertainment. 

With many mistakes and some successes, at grips with their difficul- 
ties and intimate with their miraculous achievements, I have learned 
that no story ever written for the screen is as dramatic or as romantic 
as the story of the screen itself. It is rather a privilege, you know, as 
well as a responsibility at times overwhelming, to have such a part in 
providing the principal and essential amusement of all the people. 
I am grateful for that privilege. And I am grateful, as all must be 
grateful, for the immeasurable contributions which you and your 
craft have made. To say immeasurable does not overstate this 
service. 

The pace of progress is slow. It seldom moves at a thunderous 
gallop, but sure-footed, it circles the earth with short, unending 
steps. Thus you have worked in field and laboratory. Patience and 
ingenuity have developed the intricate and complex processes which 
have given us action, sound, color, the effect of light and shadow the 
instruments which have made it possible for the motion picture to 
perform its service to art and industry, and become the greatest 
universal entertainment the world has ever known. Your efforts will 
result in developments of tomorrow, challenging the writers, directors, 
and the artists to employ their talents to the utmost to make full 
use of the infinite resources you put at their disposal. 

I like to think of our present inventive and engineering factors as 
disciples of the master of your craft, and I commend you now in that 
spirit. A thousand years from now men will revere the name of 
the man who also gave us this really priceless gift. I would salute 
him now as the master miracle worker of all time, the man who made 
the dreams of the world come true Thomas A. Edison. 

It is the very nature of the art that pictures reflected from the 
screen should be ever moving toward wider fields of human service. 
Dynamic in its appeal, the screen can not long be standardized into a 
single art form. It is the one art-industry, the very essence of which is 
motion, sound, and color, vivified into the nearest possible representa- 
tion of life. Only a few years have elapsed since the moving shadows 
have received a voice, and the utterly silent picture of yesterday seems 
like a poor crippled thing, stumbling along in speechlessness, des- 
perately trying to make itself understood by placards and pantomime. 
The silent film depended for its impression upon external conduct 
and behavior upon action. What the screen players did could 



Dec., 1935] PROCEEDINGS OF SEMI- ANNUAL BANQUET 485 

easily be portrayed; what they thought, the audience learned only 
through indirection, or through printed captions. Today it is 
possible for the camera to photograph mind, as well as movement, 
and to reveal the intellectual and emotional interpretations of life 
and literature. The subtleties of psychology and the drama of 
human motivation can now be shown upon the screen. 

We must remember that motion pictures are at once an art, a 
science, and a business. From the very beginning our technical and 
artistic progress came hand in hand. It was essential at every step 
that art wait upon science. For, after all, the elements of drama have 
not changed materially in three thousand years. We are still build- 
ing upon the original dramatic situations. In essence, Elizabethan 
drama, barring costume, stage effects, and scenery, is not vastly 
different from the Broadway efforts of today. When this new 
medium of expression flickered into the public's consciousness thirty 
years ago, film entertainment had to deal with A, B, C's of the art 
with action that had to speak louder than words. Thus the early 
stars dived into raging rapids. They fled from comic policemen. 
They chased villains across the plains. In those early days, it was no 
wonder that those who saw the then technical limitations of the 
"movies" envisioned a perpetual babyhood for motion picture art. 

The problem of the screen has been viewed from many angles. 
Some saw this problem in a single dimension. Make better pictures, 
they argued, and better audiences will support them. They did not 
recognize the dual process of development which required both con- 
stantly improving standards of production and higher standards of 
appreciation. They did not know that improved supply and im- 
proved demand were twin necessities, equal one with the other. 

We of the industry, who must face the problem from all standpoints, 
know that to advance steadily, motion pictures must proceed in three 
parallel lanes. There must be technical development by which the 
screen may enlarge its artistic field; there must be better pictures 
from the dramatic as well as from the social standpoint; and there 
must be standards of public appreciation by which the right pictures 
will find the right audiences. These movements must be simul- 
taneous. We can not lag upon one road without stopping progress 
upon all three. Without the technical development that went side 
by side with artistic progress, talking motion pictures could not have 
reached their present artistic merit. 



486 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. p. E. 

We are now reproducing, in the form of high entertainment, the 
great spectacles of history and the vivid chronicles of our own country. 
Too, pictures are moving into higher spheres of dramatic and educa- 
tional appeal. The severest critics of the movies, from the social if 
not from the artistic standpoint, have been loud in their praise of 
recent films based upon fine, wholesome drama, upon entertainment 
that enriches human understanding. We have now reached the 
stage where the great music of the opera, the symphony, and the 
concert hall, dramatized in the universal entertainment that de- 
lights both the eye and the ear, is being brought to millions. 

In the past year, in addition to its treatment of original and current 
themes, the industry has vivified great works of classical and current 
literature into outstanding film entertainment. All this trend is on an 
upward curve, as evidenced by A Midsummer Night's Dream and 
such forthcoming productions as A Tale of Two Cities, Peter Ibbetson, 
The Three Musketeers, Dodsworth, The Good Earth, Green Pastures, 
Ivanhoe, Kim, Marie Antoinette, Mary of Scotland, Pickwick Papers, 
Romeo and Juliet, Hamlet, Silas Marner, The Life of Beethoven, 
Under Two Flags, Pasteur, Captain Courageous, Dr. Samuel Johnson, 
Ramona, General Grant, Twelfth Night, As You Like It, Last of the 
Mohicans, Valley Forge, Suiter's Gold, Sam Houston, Courtship of 
Miles Standish, Samson and Delilah, Way Down East, and many 
others. 

This list sounds a little like Dr. Elliott's Book Shelf yet it is part 
of the projected program for mass entertainment; and it is, indeed, 
mass entertainment, for not less than ten million persons per day see 
motion pictures in this country alone, and not less than 27,000 miles 
of film are handled each day in our exchanges. That list, and the 
pictures that are coming it will be hard to deny them. 

Toscanini had once concluded a great performance. The hall rang 
with thunderous applause. The concert was so successful there were 
expressions of mutual appreciation in the orchestra. The first vio- 
linist alone was sour, standing with wry face. So obvious was his 
distress that Toscanini, inquiring as to the trouble, said: "Pedro, 
didn't you like the selection of the program?" "Oh, yes," said the 
violinist, "I thought it was perfect." "Well, Pedro, didn't you like 
the way I read the score?" "Yes, Maestro, it, too, was perfect. 
No one could have read it better." "Well, Pedro, didn't you like the 
way I conducted?" "Oh, Maestro, it would be sacrilege to suggest 



Dec., 1935] PROCEEDINGS OF SEMI- ANNUAL BANQUET 487 

otherwise. No one is as great as Toscanini." "Well, then, in 
Heaven's name, Pedro, what was the matter?" Pedro, with another 
wry face, exclaimed, "Ah, Maestro, I just don't like music." 

Such combined entertainment and cultural service to millions 
would not be possible if, in addition to the will, there were not the 
means to accomplish them. To develop the mine of literary, dra- 
matic, and musical material, the screen not only had to have a voice, 
but had to learn how to use it. The marked achievements in techni- 
cal development argue conclusively that research, resourcefulness, 
and invention will continue to underline further progress. The 
talking picture is only at the beginning of its career, artistically and 
socially. Our studios are the workshops of authors, scenario writers, 
directors, musicians, artists, and producers, as well as the laboratories 
for scientists. 

And what workers they are ! I sometimes think that to do the job 
as they do it they must have all been fed the spinach of Pop- Eye, 
the Sailor. From centuries of study of light and sound, from the 
profundity of pure science, they have builded a peak of sheer art as a 
platform upon which to present the monkey-shines of Mickey Mouse, 
and for him they have won wild applause from the learned; even as 
they have brought from the masses acclaim loud and long for the art 
of George Arliss, as Richelieu. Paradoxical. It simply can't be 
done. But there it is. The miracle of the movies. The most sig- 
nificant social phenomenon of the generation. 

Like the printing press, the invention of the camera marked a 
fundamental stage in the march of human knowledge. The motion 
picture camera which brought to life the characters photographed was 
another development. Improved technic has given full rein to the 
skill of our camera experts. Our technicians have performed miracles 
of illusion. But such advance will never stop short of its true goal 
which is the closest possible imitation of nature and life. Natural 
objects have three dimensions. So I am confident that the audience 
of tomorrow will witness pictures with the qualities of natural per- 
spective, height, width, and depth. 

Thrilling as a baby's first cry was the event which announced that 
shadows could talk as well as walk. But no less important are the 
later achievements of sound recording and reproduction. Artistry, 
at last, has learned to use speech and music effectively in the creation 
of talking motion picture entertainment. Too, the screen's de*but 



488 PROCEEDINGS OF SEMI-ANNUAL BANQUET [J. S. M. P. E. 

into the field of great music, not merely as a background for action, 
but as a part of the main entertainment theme, is as significant 
perhaps as was the fact that at last it had received a voice. 

It is only by the creation of the better, the truer, the finer, that we 
can measure the progress still possible. Men were thrilled by the 
fact that their first motor cars reached a speed limit of twenty miles 
an hour. They sat riveted to radio sets that hissed and crackled and 
faded. They were content with the crudest stage devices until 
greater artistic progress was achieved. Sufficient for the day was the 
good thereof. The victories of greater screen illusion, both in sound 
as well as sight, are before, and not behind, the art. 

As with sound, so with color. Color plays an important and con- 
tinuous part in our lives. It is another element of vivid dramatic 
presentation. It is a stroke of realism which eventually must mean 
much to motion picture appreciation. We shall learn not only to 
achieve the highest form of naturalness in color and color combina- 
tions, but also the most effective manner of their use. 

Artistically as well as technically the film is growing in the im- 
portance of its appeal. Every form of dramatic expression finds its 
outlet today in modern talking pictures, and they are coming to re- 
ceive their measure of artistic recognition growing to a high artistic 
and esthetic stature from an infancy whose promise few could dis- 
tinguish. In a single generation a world public has witnessed the 
development from silent flickers of motion to a great new art of the 
theater, where all that is fine in drama, literature, music, painting, 
and sculpture may be reproduced in motion, sound, and color. 

But the industry can not and does not rest upon its artistic and 
technical success. Underlying the advance of the screen is a con- 
tinuous process of education, translated in the form of self -regulation 
within the industry and public cooperation without. It would 
matter little what science and technical progress could create, what 
artistic genius could achieve, if the right kind of pictures did not find 
the right kind of audience. Motion pictures are an industry as well 
as an art. Universal entertainment demands that the screen be 
within the reach of the vast majority of our people. The industry 
can not outrun with impunity the requirements of the public which it 
serves. But it can, should, and must strive for ever-higher forms of 
entertainment appeal. It is, indeed, a continuous educational task, 
in which public cooperation and constructive criticism must play 



Dec., 1935] PROCEEDINGS OF SEMI- ANNUAL BANQUET 489 

their vital parts. This is not the occasion to discuss the many pro- 
cedures of self-regulation which we have developed during the last 
fourteen years. The sum and substance of such procedures must be 
reflected upon the screen. Pictures, not words, must tell that story. 
The technical, artistic, and social progress has a very definite bearing 
upon the educational and cultural aspects of the screen. 

I have on a previous occasion discussed before your body the im- 
portance of collecting and preserving the picture records of historical 
occasions. I have called attention to the value of sound, action, and 
color that would preserve significant contemporary events for the 
coming generations with the vividness, realism, and certitude of life. 
Through no other means can the pageant of history be recorded in the 
living tempo of the time in which the events occur. 

The American motion picture industry has produced, and has, 
countless miles of newsreels and historical subjects reporting the 
outstanding events and picturing the great figures in the inter- 
national arena during this, the most stirring period, perhaps, of world 
development. Moreover, the entertainment screen is constantly 
adding highly artistic and faithful reproductions of the American 
scene as described in the records of history. Some progress has 
been made since I emphasized this need at your meeting on May 7, 
1930, but it would be a gross injustice to posterity if we failed to 
organize this moving, living, talking record so that historical and 
educational material might be available to the students of the future. 

Films are printed upon celluloid. These are records infinitely 
more perishable than the records carefully tended in libraries and 
educational institutions. Many priceless chronicles of current and 
contemporary events are literally fading away for want of careful, 
scientific preservation, which must follow a system of collation and 
selection of films from a vast footage of material produced every year. 

Therefore, I commend the purpose of a Film Library in the Na- 
tional Archives Building in Washington, as an important step in the 
right direction. The full task will require the coordinated efforts of 
many agencies, public and private, and I reiterate the desire of the 
industry to give full cooperation. To this end, it is willing to con- 
tribute its facilities and its expert knowledge. 

We have now reached the stage that clearly warrants the extension 
of the motion picture art into the field of teaching. Through its 
technical, artistic, and cultural development, the film has opened a 



490 PROCEEDINGS OF SEMI- ANNUAL BANQUET [J. S. M. p. E. 

new vista of educational possibilities. Sound and color have added 
vastly to its range. 

A great field of service calls for the joint cooperation of science 
and education, on one hand, and technical development by the 
industry, on the other. There is no substitute, in the field of visual 
demonstration, for modern motion picture photography synchronized 
with sound. There is no substitute for the graphic appeal to im- 
agination and memory which the talking picture makes. "Visual 
education" may have expressed the limitations of the old art, but a 
wider term is necessary where we can combine sound, color, and 
motion in the service of teaching and learning. Every step of prog- 
ress which the motion picture industry makes brings us nearer to the 
day when text -books will be written upon celluloid as well as upon 
paper and the value of this supplemental tool for the teacher is demon- 
strated'. The educational community will owe much to the pioneers 
who have been and are seeking to make this new medium effective 
both in adult and child teaching, in the lecture halls of the Universi- 
ties and the classrooms of our schools. 

With progress such as this, we may well envision a screen in which 
even closer realism, molded by the basic principles of art, will mark 
the great motion pictures of tomorrow. Finer harmony will char- 
acterize every phase, and the widest possible service of entertain- 
ment, information, and education will obtain. 

We may imagine the visual surface of a screen animated with a 
stereoscopic background, with every natural color undiminished. 
We may face, in a great rendition of opera, a scene more impressive in 
its vividness and naturalness than an opening night at La Scala or 
the Metropolitan, taking the finest artistry of the world to every town 
and city in the nation. 

We may foresee entertainment which will continue to draw upon 
new story values of high dramatic intensity, both from life and from 
literature. But progress will not be limited purely to dramatic fea- 
tures. The newsreel will achieve an even greater place in the coun- 
try's informational services, ranging the world in trail of significant 
events. Short subjects will continue to show one half of the world 
how the other half lives, works, and sings. The teaching film will 
become a standard medium of education, and not a novelty in the 
classroom. So far, indeed, has the science of the motion picture 
advanced, technically and mechanically, so surely has its art quality 
been demonstrated, so wide is its scope, that its present possibilities 



Dec., 1935 J PROCEEDINGS OF SEMI- ANNUAL BANQUET 491 

constitute a tremendous challenge to those who would create some- 
thing of beauty or of lasting value. 

This progress is a challenge to the teacher to discover the films 
maximum usefulness, and to adapt it to his field of education. It is a 
challenge to those groups which seek to advance good taste and social 
welfare. It is a challenge to science to apply the film's capabilities 
to the full. It is a challenge to the artistic mind and hand to advance 
their purpose with the infinite means now placed at their disposal. 
It is perhaps the greatest challenge of all to the writer, and the pro- 
ducer, and the director, because the screen's major purpose is enter- 
tainment. 

If imagination can project a humanity deprived of its entertain- 
ment release, all screens silent and lifeless, then it will conjure the true 
measure of the vast and vital benefit of the motion picture. It 
satisfies a deep and fundamental craving as basic and human as the 
desire of a hungry man for bread. Great civilizations of the past 
perished amid the fanfare of distorted entertainment standards. 
The motion picture is working in historical contrast to those de- 
cadencies, profound and far-reaching. The continued evolution of 
the screen to higher forms of entertainment may be one of the strong- 
est fortifications of our own civilization. 



ANALYSIS OF THE DISTORTION RESULTING FROM 
SPROCKET-HOLE MODULATION* 



E. W. KELLOGG AND H. BELAR** 



Summary. That constancy of record speed should be sought in sound recording 
and reproduction has been recognized throughout the history of the art, but standards 
of performance have usually fallen far short. Especially have the speed variations 
due to sprocket-tooth action failed to receive from many engineers in the sound-picture 
industry the consideration which their harmful effects warrant. This is due to imper- 
fect understanding of inherent limitations and to the fact that the impairment of 
sound quality is of a kind which may easily be ascribed to other causes. An analysis 
is made of the loss of true tones and production of spurious tones, as a result of speed 
modulation at sprocket-hole frequency. A recorder in which not only are disturbances 
of this kind eliminated, but all speed variations reduced to a minimum, has been 
available to the film recording industry for several years. Film phonographs, re- 
recording machines, and projector sound heads of which this is true are now also avail- 
able, thus making possible an over-all performance which sets a new standard in the 
sound reproducing art. 

Ever since the beginnings of the art of recording and reproducing 
sound, constant speed has been recognized as a desideratum. Edi- 
son's first phonograph employed a large flywheel and, in the discus- 
sion in his U. S. Patent, No. 227,679 (May, 1880), he showed a very 
clear understanding of the importance of speed constancy. Many 
other pioneers in the sound recording art have either shown expedients 
for improving the speed constancy of their devices or else, without 
showing any expedients, have postulated "constant speed." Al- 
though it was realized that perfect functioning of a recording and 
reproducing system called for constant speed, an approximation to 
constancy has had to suffice even to the present day, both in phono- 
graphs and in sound-picture equipment. In the phonograph field, 
commercial considerations permitted the acceptance of standards 
which have not represented the best results attainable. Consider- 
ably more effort, however, has been devoted to speed constancy in the 
case of recording machines than in reproducing machines, since only 

* Presented at the Spring, 1935, Meeting at Hollywood, Calif. 
** RCA Manufacturing Co., Camden, N. J. 

492 



SPROCKET-HOLE MODULATION 493 

small numbers of machines are involved, and the defects of these 
machines are impressed upon all the records made. 

In the sound-picture field the fact that so much equipment of 
inferior performance has been developed has been due to several 
causes. There has been much misunderstanding of the problems; 
and, in particular, the harmful effects of sprocket-tooth action have 
not been fully appreciated. This has been the more true because the 
effect upon sound quality is of such a character that it is readily at- 
tributable to other causes. The presence of many other faults and 
sources of distortion has diverted attention from the speed modula- 
tion problem, and standards of performance have been accepted 
which did not represent the best attainable. The faults in reproduc- 
tion due to speed irregularities are more obvious with music than with 
speech, and the fact that the sound equipment became largely a 
vehicle for speech rather than for music has not stimulated the 
adoption of the best possible designs. 

In the pioneer development of Photophone apparatus at the 
General Electric and Westinghouse Companies, speed constancy re- 
ceived more attention than it might otherwise have received for the 
reason that the function of sound equipment was at first conceived 
to be for musical effects quite as much as for speech. The imperfect 
reproduction of piano music as a result of speed fluctuations was 
subject to immediate criticism and a number of different types of 
film-moving systems were evolved in an effort to find the best so